Cohort Directory

Nitzsche

Cambridge, MA

Hatton Research Group - Massachusetts Institute of Technology

My work is in electrochemically driven carbon capture for decarbonizing heavy industry, and I am hoping to start a company to mitigate hard-to-abate emissions. I'm passionate about sustainability and the environment, and my career goal is to maximize my negative carbon footprint.

Our technology uses an electrochemically mediated amine regeneration process (EMAR), where plating and stripping of copper ions into an amine solution drives capture and release of CO2. Initial work on EMAR began in the Hatton Lab nearly a decade ago. Several students and postdocs have worked on the technology, with patents and papers ranging from fundamental chemistry to system-level design, and demonstrations reaching the 50 gCO2/day scale for 100 hours. Michael and I have continued to develop the technology in the lab, with our recent IP advances focused on de-risking and scalability for commercialization.

On the technical side, de-risking efforts will focus on understanding and enhancing long-term stability, particularly in the presence of oxygen and improving chemical kinetics, as these can still be done at the bench scale. Towards commercialization, our primary goal is to improve our system-level technoeconomic models with benchmarking against existing carbon capture technologies to assess commercial viability and find our best market fit.

Amin

San Francisco

Stanford University (for now!) I’m leaving to go full time on founding Collage Biosciences later this year.

I am developing RNA splice-modifying therapeutics for neurodegeneration and the aging brain. By developing splice-sequencing technology and an ML-driven computational biology approach, we are building the first atlas of brain splicing across the lifespan that will power our target detection platform. By applying my organoid screening platform and in vivo preclinical models, we will improve neural health and develop our lead candidate - a antisense oligonucleotide drug modifier of neurometabolism. We are aiming for first in man clinical trials within 5 years.

I’ve spent the last 16 years studying medicine and neurobiology through my physician scientist training. My work in RNA processing the nervous system and development of organoid technology has lead to publications in nature, cell, science, nature medicine, neuron and nature neuroscience and patents from Stanford and the Salk Institute. This company idea has been honed over this time and is the culmination of my scientific obsessions and drive to improve brain health, prevent disease, and challenge conventional biomedical wisdom around what is possible.

My biggest current risk is lack of time with my full time faculty position at Stanford. I am leaving next month to stop dreaming and start building as full time founder. Next is honing the clinical stage strategy and mapping an FDA approval plan that will guide our efforts. I will actively fundraise to continue supporting my current bioinformatics tram that is transitioning from academia. Concurrently I will establish a lab and develop technology for our first patents around splicing sensitive RNA sequencing. The bioinformatics team will begin lead candidate identification efforts with our available data while we obtain aging and disease brain samples to generate proprietary data.

Mooney

Cambridge, MA

Postdoctoral Associate

I develop low-cost decentralized water technologies, using the moisture in the air and sustainable heat (solar waste or waste) to produce this water. We can operate in the most arid of climates in either a passive or active manner. I see myself building a device to be ready for pilot in the next year, looking for immediate funding to incorporate, establishing myself in a climate tech hub, hiring my first employee, and developing a plan for the pilot. I would envision my first customer within the agriculture space (small-scale crops or hydroponics).

I have worked with researchers to demonstrate record water uptake performance with our materials. I have developed three technologies/ devices (two passive and one active, and two at different scales), tested two on the roof of MIT and show record amounts of water produced, and tested one device in the driest non-polar desert on Earth (the Atacama desert) and also shown record amounts of water produced. Our next mission is to fabricate it at a larger scale. I have done the geospatial mapping of weather data to show a country's need for the technology. I have also tried to build myself as an entrepreneur by attending internal accelerators.

There is a market validation risk (i will test a full scale device and couple it to a small set up driven by the customer), there is a risk in making our device autonomous and powered by a small PV (this will be de-risked by running a device for 2-weeks off the grid and recording data), there is risk in securing funding (I will start approaching investors, working with academic collaborators and attending start-up competitions to generate funding), there is risk in terms of my current visa expiring in February 2025 (once I secure the funding I will apply for a GEIR visa in UMass or something similar, I will also apply for a green card EB1-12, if I cannot get a visa on time I will convert to a H1-B research specialist in my lab) and funding would help support spin out and a visa, there is a risk in not convincing the customers/ investors (I would love to work with 50 years to build myself to communicate the technology most effectively and do due diligence in identifying that first paying customer).

Zhang

Mountain View

Apple

leverage data to systematic design battery materials and manufacturing

understand the gap between data-driven insights and experimental results

Cao

Boston

Carnegie Mellon University / Boston Dynamics AI Institute

Autonomous mobile robots that are useful in people's daily life

Building a physical robot and surveying people on how the robot could be useful

Talk to potential customers and users to determine if there is an actual need and market for the robot.

Liu

Boston/New York

MIT

I work on AI and its applications for science domain. My vision is to build AI tools designed to accelerate the screening and proposal of new drug/material development, using the idea of AI agent interacting with multiple AI models.

A novel AI method and experimented it on standard image/text benchmarks.

Figure out its best use case for science discovery and the best product-market fit.

Venkatramani

Boston

Harvard University

I am building a device to measure tissues' RNA/ DNA/ protein content at single-cell resolution with genome-scale sensitivity. My vision is to enable the creation the usage of genomic data to understand organs and organisms in their native or abnormal states.

I have built a proof-of-concept machine that can measure RNA content with high spatial resolution and with the promise of high genomic-scale sensitivity. It also has the promise of scaling up the number of cells we can measure but that is yet to be demonstrated.

On the scientific front, I am working on scaling up to make sure this will work better than existing tools in the market. On the spin-out front, I am working on figuring out ways around IP conflicts. I am also exploring avenues outside of building a device company such as thinking about applications in precision medicine or working with tissue banks. This is very much work in progress and want to try to stay true to my bigger vision.

Burger

Boston, MA

Chouchani Lab, Dana-Farber Cancer Institute and Harvard Medical School. Previously, Murphy Lab at the University of Cambridge, UK

Sleep is a major determinant of longevity and protective against metabolic, cardiovascular, and age-related diseases. However, little is known about the molecular mechanisms underlying sleep. I propose that metabolic adaptations during sleep regulate cellular functions and thereby convey health promoting and pro-longevity effects. Defining the metabolic principles of sleep will provide us with a unique therapeutic handle which we can leverage to develop a new class of therapeutics to target and activate the pro-health mechanisms of sleep. Harnessing sleep will revolutionize the way we age.

I have recently completed my work to generate a global atlas mapping the evolutionary trajectory of all 260.000 human protein cysteines across close to 3000 species. I further enriched the dataset with structural data and additional information to establish a unique profile for each cysteine which informs on their reactivity and regulatory potential. I am currently working full time to identify cysteine targets that correlate with requirement for sleep across species, which will be subsequently validated to identify their functional role in mediating the health promoting effects of sleep. Following biological validation, covalent drug screens will be performed for high value cysteine targets aiming at the development of potent and selective modulators of target pathways.

The first critical step will be to identify regulatory cysteine targets which correlate with the requirement for sleep. Alternatively, conserved cysteines (independent of sleep requirement) might be differentially regulated depending on sleep requirement which would necessitate in vivo studies to identify regulatory signatures in form of cysteine modifications, which is substantially more laborious.

Next, in vivo work in rodent models will be required to establish sleep dependent regulation of identified targets.

Further validations will need generation of mutant animals.

Magen

San Diego, CA

Voyant Bio

I want to build a system to enable every cancer patient (and beyond) to reach a cure instead of prolonging lifespan in just a few months with current treatments.

I have done research in the immuno-oncology space and developed a prototype platform to discover how immune cells talk and drive response or resistance to current therapies.

Partner with clinics to get patient tissue

Complement the team with advisors and early employees on the diagnostic/therapeutic R&D and commercialization

Generate additional validation/discovery data

Lay the foundations for partnering/selling to healthcare insurance or pharma companies

Raise dilutive and non-dilutive funding

Gehring

Berkeley, CA

Streets Lab, UC Berkeley (previous Latent Labs, Arcadia Science, UW, Caltech, UC Berkeley)

I develop advanced technology to make discoveries in biology. I have always seen biology as a data-limited field, and now I'm looking to combine generative modeling with high-throughput and pooled experiments, using cutting-edge data and modeling to bring new therapies into existence.

I'm implementing and modifying deep generative models for protein design. I have developed advanced experimental and computational methods for single-cell and spatial genomics, and I am currently developing a next-generation spatial genomics method that could screen computationally designed biologics at scale. I am exploring application areas for biologics, but I have not yet settled on a target disease or therapy. I have crossed off small molecule therapeutics in favor of proteins, and I'm focusing in on a viable screening and development cycle for designed protein therapeutics.

1. Establish computational pipeline for protein design

2. Proof of concept experiments for next-generation spatial omics

3. Hone in on target disease and therapeutic strategy

4. Integration: articulate a development cycle that applies generative modeling and laboratory screening to the chosen therapeutic area

Zhang

Boston, MA

PhD Candidate in Biological Engineering at MIT

My research is focused on HIV vaccines and cancer immunotherapy. I wish to build cutting-edge biotechnologies that harness our immune system to improve human health.

I have done extensive market and background research and put together a pitching deck.

I will talk with more expertise in various fields to gauge people's interest and skepticism.

Kenny

Seattle

Baker Lab, Institute for Protein Design

I design proteins to perform novel functions that our current therapeutic strategies can't achieve. I envision a world where no diseases are not treatable.

I have shown proof of principles the therapeutic strategy I am trying to commercialize

I filed the patent for this strategy through the university's tech commercialization office. I am planning to develop the business model and start the licensing process during the 5050 program timeline.

Medetgul-Ernar

Stanford & Teton Valley, Wydaho (WY + ID)

Stanford MD-PhD Program with PhD in Biophysics at Mark Davis Lab

Translate our new blood test into the clinic, establishing the new common blood test.

Developed it and tested it across multiple immune conditions

I want to attend as many 50Y events as I can and learn as much. I want to ship this test for everyday people (immune health, inflammation, and other nonclinical uses) so that I can start selling if I need to - and gather data.

van Kooten

Palo Alto, CA

Stanford University

I am on a mission to reboot the human powerplant. Advances in omics measurements, mRNA technology and genome engineering offer unprecedented precision and potential to deliver specific genetic instructions to cells to restore or enhance function. I am an engineer, and these cutting-edge tools make it possible to directly target and repair mitochondrial dysfunction, offering a real opportunity to redefine human health span and lifespan.

Structure: We set out to build Powerhouse Biology as a Focused Research Organization, a concept that we applaud but one that is simultaneously complex to achieve in a controllable manner – the concept heavily depends on philanthropy - and to build up in a milestone-based approach. Therefore, I recently pivoted and am now shaping Powerhouse Biology into a to-be commercial entity. Science: The initial aim was to move the mtDNA into the nuclear genome. Since, I have shaped this aim into concise and commercially exciting milestones and in R&D, focus on these milestones. Viability: I am currently focusing on generating IP.

I am currently focusing on generating IP, which on a day to day basis involves research and making good decisions in research. In addition, I want to further explore a plan B re IP should the science not work out, settle on a co-founder, and shape the business case.

Seong

Berkeley, California

Ksenia Krasileva lab at the University of California, Berkeley

Sinha

Boston

Ed Boyden Lab, MIT

My goal is to systematically map the involvement of the peripheral nervous system in nearly every disease, and to design novel therapeutics that target this missing puzzle piece of human biology.

I am landscaping the state of the field across diseases (1) to identify novel disease areas where there should be an interaction that hasn't been discovered yet; and (2) to identify classes of mechanisms that are shared across diseases, to be able to clearly articulate what a bioplatform company would work on.

Kumar

New Haven, CT

CS PhD @ Yale working on quantum computing, advised by Yongshan Ding. Fully funded by an NSF fellowship. Prev. @UChicago (Undergrad), Oxford Physics (Research), and Super.tech (Quantum Computing startup now acquired by Infleqtion). Will be @MIT in Spring 2025, working on ML for quantum computing.

I'm leading multiple projects on the topic I described above. That being said, I am still at a very early stage in terms of (1) research/potential IP and (2) figuring out how to commercialize it. My approach to quantum software is based on my work at Super.tech, which also built a quantum software stack relating purely to QEM techniques and subsequently got acquired. I have not iterated more on that fundamental idea in relation to todays' quantum industry and my own niche.

My plan is to focus solely on the fundamental research and de-risking the technology for now. Maybe that will change after 5050.

Tu

Boston

MIT

We are building a transformative technology that measures protein activity in ultra-high throughput, which enables AI-driven engineering of next-generation therapeutics and industrial enzymes

We have designed and tested the core components of this technology. We were able to generate millions of data points, train ML models for denoising and show how our data can improve protein function prediction.

We are going to leverage the data we have and millions more data points to be generated in the next month to engineer TEV protease to specifically recognize an arbitrary substrate. This will allow us to demonstrate the broad utility of this technology in protease engineering.

Sundar

Boston

Sculpting Evolution Lab (Kevin Esvelt), MIT

We have developed a novel high-throughput assay DHARMA that measures protein fitness, and a novel Bayesian denoising technique called FLIGHTED to denoise the output of a DHARMA assay. Combining these, we have measured the fitness of large datasets of TEV protease variants as a first step towards engineering TEV protease.

Scientifically, we want to show that we can engineer TEV protease successfully to cut a different substrate. Ideally, we would like to also demonstrate specificity and that TEV protease can be engineered to cut an arbitrary substrate (i.e. without experimental data specific to the given substrate).

Szedlock

Stanford University (Palo Alto, CA)

Zheng Lab, Stanford University (PhD student). Previously worked at Soaring Co (drone startup based in Southern California) as a lead mechanical engineer. BS in Mechanical Engineering from Caltech.

I currently work on developing methods to make lithium-ion battery recycling cleaner, more efficient, and more economically feasible through combustion-based pyrometallurgy. By using parts of battery cathodes as in situ reductants, we can more effectively recover precious materials such as lithium and cobalt without harmful carbon additives and reduce the need for toxic solvents. My goal is to help create a future where lithium-ion batteries are part of a fully circular economy which reduces our reliance on mining precious metals and makes electric vehicles and other forms of stationary energy storage far more affordable. I hope that this will help accelerate the green transition and unlock a number of other technologies which are currently unprofitable or supply-chain limited.

During my PhD, I have worked on characterizing and developing a process that uses aluminum as an in-situ reductant to improve the recoverability of lithium, cobalt, nickel, and other precious metals. This method takes a fraction of a second and the reaction itself is self-sustaining with significant heat release which shows promise for repeatability at larger scales. Through methods including XPS and XRD, I have shown that we can reliably reduce the oxidation state of our precious metals which makes them more leachable in organic solvents compared to untreated cathode powders. Our recovery rates are currently in the 80-90% range and our method requires fewer pretreatment steps compared to standard commercial processes with less carbon emissions.

My plan for the next few months is to fully develop the processes needed to replicate this method at scale and to ensure that it reaches commercial-grade repeatability. I also want to conduct a thorough techno-economic analysis of my process and compare it to existing industrial methods to fully quantify the competitive advantage. Finally, I would like to transition from lab-scale experiments towards breaking down larger scale batteries and applying this method to see how well it can perform with battery scraps and other recyclable lithium-ion batteries.

Viala

Boston

Tesla, Imperial College

Al-Shimary

University of California Berkeley

Doudna and Savage Labs UC Berkeley

Our primary focus so far has been on the development and refinement of the tool and pipeline. We have expanded the range of targeted phages and demonstrated that we can effectively target the full spectrum of bacteriophage encoding strategies—something no other tool can claim. Through this process, we have gained a clear understanding of where the tool excels and where it falls short, enabling us to create efficient workflows.

Weiss

Cambridge, MA

MIT, Brushett Lab PhD Candidate (5th year)

My work focuses on developing advanced lithium-ion battery cell architectures that address many of the current limitations in such battery technology. Rather than relying solely on new materials, our approach leverages an engineering perspective to overcome fundamental issues related to mass and thermal transport. This strategy has the promise of enhancing both the performance and lifespan of lithium-ion batteries, making them more effective and reliable.

Our goal is to accelerate and expand the adoption of energy storage solutions for electric vehicles and grid applications while unlocking new possibilities for battery use in sectors like trucking and drones. We envision a future where energy is more sustainable, accessible, and reliable, with solutions that seamlessly integrate into modern infrastructure to create a more resilient and adaptable energy ecosystem.

We have made strong progress in both scientific and commercial de-risking of our battery technology. Initially, we focused on developing and validating the core principles of our engineering approach. This involved extensive research to model heat and mass transfer within our battery cells, as well as conducting system-level and technoeconomic analyses to confirm the commercial validity of our solution.

We have since transitioned into exploring different materials and assembly techniques to design alternative cell architectures. Our goal has been to find a balance between methods that leverage existing infrastructure and those that optimize performance. Currently, we are using existing methods with minor modifications to enable rapid prototyping and to develop a minimum viable product (MVP) that demonstrates our technology’s capabilities in real-world applications.

On the commercial side, we have engaged with potential industry partners and stakeholders to better understand market needs and refine our value proposition. This includes discussions with battery manufacturers, electric vehicle companies, and grid storage providers to identify specific pain points that our technology could address. These conversations have guided us in prioritizing features and applications for which our approach offers the greatest value.

Our next critical step in de-risking our technology is to finalize an initial design and conduct associated testing to demonstrate its performance in real-world conditions. This will involve lab testing and characterization, and may include advanced simulations to create a feedback loop for interpreting experimental data and optimizing the design.

Simultaneously, we plan to identify our most viable beachhead market and potential strategic partners. Engaging with these partners will help us gain insights into specific use cases, regulatory requirements, and integration challenges, allowing us to refine our product roadmap accordingly.

Once we successfully demonstrate the technology experimentally, we will look to procure initial funding to support further R&D and begin scaling efforts. This will include raising seed investment and applying for grants that align with our mission.

Lal

Pasadena, CA

Currently I am a graduate student in the Arnold lab at Caltech. Previously I performed an internship at Google X. I performed undergraduate research in the Keasling lab at UC Berkeley.

I am a PhD student in the Arnold lab at Caltech working on the evolution of enzymes to improve their capability to perform synthetically challenging chemical transformations. My research currently is most focused on the development of wet lab protocols which generate protein sequence-function data efficiently for integration with machine learning methods which could be used to accelerate the directed evolution process. I hope that this research and any subsequent research that I perform will play a part in the growth of the bioeconomy. I envision a future where many of humanities chemical needs have been made more sustainable as our understanding and ability to control biological systems improves.

So far we have worked on projects in the Arnold lab involving the use of novel machine learning methods for the acceleration of biocatalysis-focused protein engineering. We are focused on learning the bounds of these processes during our time here. Some areas we have crossed off are de novo protein design, antibody engineering, and projects which are solely based in the dry lab. We have also done a cursory evaluation of current biotechnology companies to evaluate where there are opportunities in the bioengineering space.

Our plan for the next few months is to utilize our engineering methods and expertise to arrive at a marketable product which will benefit society by providing greener methods for chemical synthesis. A secondary goal of ours is to ensure that the product we develop serves the needs of the chemical community by providing access to more sustainable and robust catalysts. One de-risking step we plan to take is to talk to potential customers to help us arrive at a desirable product.

Zhong

Cambridge, MA

Device Research Lab (led by Evelyn Wang), MIT

The current water infrastructure is unsustainable – it contributes 2% of global greenhouse gas emissions, and yet 80% of wastewater is discharged untreated due to a lack of cost-effective solutions, posing a serious threat to our environment and health. I want to fill this gap by turning wastewater into freshwater, clean energy, and fertilizer, transforming the existing water infrastructure.

My research at MIT focuses on a membrane-based, low-temperature distillation technology platform for efficient liquid-vapor separation, applicable water, ammonia, and methane. Leveraging my expertise, I want to develop a scalable unit that converts wastewater and carbon dioxide into increased electricity, heat, and fertilizer, while producing freshwater. I have done some preliminary market and technoeconomic analysis to validate this idea.

I aim to develop a more comprehensive TEA model to validate my hypothesis and identify key cost drivers; I plan to apply for several small grants to support validation at the component level and demonstrate proof of concept for full system integration; Additionally, I’m seeking a technical co-founder with expertise in water treatment or a business co-founder with industry experience.

Sicinski

Pasadena, CA

I received my PhD in Chemistry from Tufts University in 2021. My thesis focused on developing novel chemical methods to enhance the metabolic stability of therapeutic peptides in the GLP-1 peptide family, which includes hormones like Ozempic. This work resulted in a start-up company Velum, Inc founded by my PhD advisor Prof. Krishna Kumar. During my PhD, I had the privilege of representing Velum at the Tufts University $100k New Ventures Competition, where we secured 3rd place. After my PhD, I worked as a scientist for a start-up biotech company, PepGen, Inc. At PepGen, I saw the company go from series C fundraising to its initial public offering in 2022 and advanced into clinical trials for the treatment of Duchenne muscular dystrophy. I was responsible for transitioning laboratory operations from Oxford, UK to Boston, MA and discovering new bioconjugation methods for delivery of therapeutic oligonucleotides into target cells. I then moved in 2022 to my current position at California Institute of Technology as a NIH Ruth L. Kirschstein NRSA Postdoctoral Fellow in the lab of Prof. Frances Arnold. In my current position, I utilize the powerful tools of directed evolution for protein engineering and apply machine learning methods to accelerate wet-lab discoveries. This is where I met my current co-founder, Ravi Lal. We have been working together to fast-track the iterative process of protein engineering by incorporating machine learning and high throughput analytical tools to deliver sustainable biocatalysts for the synthesis of molecules.

My current work in the Arnold lab focuses on evolving and engineering enzymes to transform low-cost chemical feedstocks into a variety of valuable synthetic building blocks. Throughout my project, I have streamlined the protein engineering process by integrating machine learning methods, enabling faster evolution of biocatalysts while reducing resource consumption. My ultimate vision is to see biocatalysts become the standard in chemical synthesis. I aim to make these engineered enzymes more accessible so that chemists can easily transition to greener, more sustainable alternatives to traditional chemical synthesis.

So far we (Ravi and I) have worked on projects in the Arnold lab involving the use of novel machine learning methods for the acceleration of biocatalysis-focused protein engineering. We are interested in learning the bounds of these processes during our time here. Some areas we have crossed off are de novo protein design, antibody engineering, and projects which are solely based in the dry lab. We have also done a cursory evaluation of current biotechnology companies to evaluate where there are opportunities in the bioengineering space.

Our plan for the next few months is to utilize our engineering methods and expertise to arrive at a marketable product which will benefit society by providing greener methods for chemical synthesis. A secondary goal of ours is to ensure that the product we develop serves the needs of the chemical community by providing access to more sustainable and robust catalysts. One de-risking step we plan to take is to talk to potential customers to help us arrive at a desirable product.

Weitekamp

Los Angeles

SpaceX

The product I envision is a low cost and high cycle efficiency Compressed CO2 energy storage system for the grid that is deployable at scale.

I'm just getting started. I've researched prior art and talked with friends familiar with this industry.

The footprint and site availability of this technology must improve in order to achieve higher market penetration. The key enabler for this is to re-invent the low pressure storage part of this system. My plan to de-risk this idea is to attack the biggest threats to viability that cost the least first (fail fast and cheaply). It is important to push the technology and market forward simultaneously. There's no point in building new technology if nobody is going to buy it.

Roy

Quebec City, QC, Canada

Former Tesla and Lyft embedded systems engineer. Currently independent.

I build robots and electric powertrains. We are facing major issues such as labour shortage and carbon emissions, where these technologies can help. I'm exploring various sectors where there are urgent needs and viable business models.

I crossed off a couple of ideas in power electronics and micromobility after running customer interviews, which (after looking back) were great solutions applied to weak problems. I'm attacking from a different angle and starting with actual big problems.

Lore

San Francisco

Buck Institute for Research on Aging

I am currently working on biology of aging research to better understand the current state of the longevity field and the current challenges of extending healthy lifespan. My vision of the world is to to see significant maximum healthy lifespan extension (at least 50%) to reduce suffering and give people more time to spend with loved ones.

I have explored many areas within aging research (ex/ mitochondrial dysfunction, genomic instability, reprogramming) and the current paths to clinic or available technologies. While I now believe that these interventions may one day help slow down aging in aged individuals or may help as a preventative intervention in young people, it seems incredibly unlikely that they will be effective at reversing damage and significantly extending healthy lifespan in individuals who are already aged. I have become increasingly passionate about replacement as an aging intervention since it can solve issues such as extracellular damage, delivery mechanisms, and irreversible DNA damage that would otherwise not be possible to fix with any other single technology or intervention.

I would like to carry out some proof of concept experiments to prove that replacement will extend lifespan and have rejuvenation effects on the brain as we would expect based on results from heterochronic parabiosis-type experiments. I also would like to continue meeting with PIs and experts in the field as well as publish an academic review on replacement in aging (I have finished the manuscript draft but still need to submit to journal) because I want to hear as much feedback as possible on why this would not work and find ways to de-risk.

Rosenthal

SF

Birdwood Therapeutics

New induced proximity modalities, beyond degraders, will revolutionize medicine. I aim to develop those modalities and unlock the ability to discover small molecule inducers of proximity through high-throughput screening

At Birdwood, we've built a team of scientific advisors and will pursue two complementary discovery platforms. Our first discovery platform enables DEL (DNA encoded library) screening to be applied for the discovery of small molecule inducers of proximity. This strategy will: - improve upon the efficiency of DEL screening to identify 'weak' (high µM) hits which could be key starting points for molecular glue discovery campaigns; - enable discovery of molecular glues for target pairs for which no known binder starting points are known; - enable screening in cellular lysates, more accurately recapitulating protein states and associational complexes; - not require protein immobilization, frequently employed in DEL enrichment screens, which may interfere with protein folding; - can detect small molecules that facilitate interactions between two proteins through intermediates. Our second platform, in collaboration with Cora Biosciences, will enable us to rapidly profile the binding kinetics for hundreds of thousands (and eventually, millions) of compounds simultaneously - a medicinal chemist's dream. This platform will supercharge drug discovery by enabling SAR studies at a scale previously unthinkable (data generation throughput > 10^6 improved). We're in the process of leasing lab space and will be ready to begin to de-risk our approach ASAP.

(see previous Q for our specific assay goals) We are planning to lease lab space to establish key elements of our platform. The first important de-risking of our technology is to show that it can be used to discover already known inducers of proximity. To accomplish this, we will use the well-studied FK506-FKBP12/FRB system (and analogs of FK506 with various kinetic/thermodynamic properties) to optimize important parameters of our technology. After showing that our technology can be used to discover already known small molecule inducers of proximity, we then plan to show that it can be used in an entirely novel context. We are still trying to figure out the best first 'test' case, but I want to go for the gold: I'd like to discover small molecules that can induce proximity between nuclear import receptors (NIRs) and proteins whose mislocalization to the cytosol causes disease (e.g., TDP-43 in many neurodegenerative diseases, FUS in FTD/ALS). These compounds can be thought of as 'molecular hitchhikers' - they cause a target protein to 'hitchhike' on the nuclear import receptor, returning them to the nucleus (and removing them from cytosolic aggregates) where they can then perform their intended function.

Katrekar

Mountain View

Currently at Arc Institute. Previously Shape Therapeutics and UC San Diego.

My vision is to make life-saving medications accessible to everyone via the creation of innovative, affordable therapeutic options. I feel like the gut microbiome can be specifically altered to produce such low cost therapeutics directly within a human.

We've gone through 2-3 rounds of iterations to identify the most achievable POC and are currently conducting scientific de-risking through experiments. I'm eager to learn how to approach commercial de-risking next.

In the coming months, the goal is to advance our scientific de-risking efforts and achieve POC. At the same time, we aim to conduct a thorough freedom-to-operate analysis and look into regulatory de-risking.

Massen-Hane

Boston, MA

MIT

I'm a Postdoctoral Associate working on new carbon dioxide capture systems that have the potential to be widely and inexpensively deployed. I'm working towards decarbonizing our power generation and industrial sectors to secure a safer and sustainable planet for future generations.

For the last 3 years, I've been analysing and developing multiple electrochemical-based carbon capture technologies in the lab at MIT. One key technology I've been working on, specifically targeted at capturing point source CO2 emissions, has been demonstrated in the lab for over 100 hours, at approx. 50 grams of CO2 per day, and with a technoeconomic assessment that puts the technology on par with the best commercially available systems. We've since made improvements to that technology, making the systems even more competitive.

As an extension to the >100-hour run already performed, I am currently determining the longer-term stability of the process when exposed to known industrial contaminants. This experimental work is being done in parallel with developing a technoeconomic model that includes parameters which account for replacing components if/when those contaminants impact them.

van de Lindt

Boston

MIT

I aim to build a company that substantially helps reduce the emissions of the maritime industry, as well as works towards ocean sustainability causes. The key technology piece is utilizing over fifty years of engineering excellence in the fission industry, as well as its new advances, and combining these with recent breakthroughs in fusion energy technology to produce a fleet of fission-fusion hybrid powered cargo container ships. This combination of technologies would play to the strengths of each: lowering the risk/physics requirements of the fusion plasma, while at the same time potentially reducing the nuclear waste of the fission system, and enhancing its overall safety.

Since a year ago, I have been working on developing the idea further in my hours outside of research with extensive modeling of the hybrid system in OpenMC, an open-source neutronics code, and plan to publish some of the results found there. In addition, for a term project I studied the economics of pure fission used in commercial shipping, and wrote a report detailing the economic advantages over alternative low-emission fuel. I took a graduate class in the marine engineering department called “marine propulsion” – going over the engineering and design of ship engines and related constraints. I was a TA and group lead for Dennis Whyte’s fusion design class at MIT for a fission-fusion nuclear waste burning tokomak, leading the fusion core design sub-team with another grad student. In a previous iteration of that same class, I was a co-author on a design publication.

I plan to aggressively improve my OpenMC model to include time-dependent analysis of the hybrid system so that lifetime reactor analysis can be done, as well as begin to build out other needed software modeling capabilities so that the concept can be refined further into a comprehensive initial design point. In addition to OpenMC, I plan to also take a deeper dive into the requirements of the fusion subsystem I have in mind. Besides engineering, I also plan to gain a deeper understanding of the regulatory burdens of this idea, and what form the partnerships would look like between my company, ship owners, and large Korean shipyards that build container ships, as well as considering what crew additions would be required (engineering, radiation, safety, security, etc.).

Gao

Boston

Brushett group at MIT

Ziesack

Boston

Current: Senior Scientist at Wyss Institute at Harvard University; Past: CTO at Circe Bioscience Inc.

I'm exploring the concept of refinery-inspired biomanufacturing, centered on three key ideas: using C1 feedstock from waste sources, super-charged bioprocessing to convert feedstock into products, and a biological downstream approach for waste-free purification. I'm also focused on partnership models and strategies for rapid revenue generation. This approach aims to create a sustainable, efficient system for producing sustainable materials from low-cost, waste-derived inputs.

On the tech side, I have built a high-level process model, written a few small grant proposals and will soon begin wet lab research funded by Wyss. On the biz side, I have had conversations with a few mid-size companies relevant to the idea; one of them is currently in NDA stage with Wyss. I've also met up and discussed my idea with two ARPA-E program directors and a few investors. I started a draft pitch deck.

Biz: Develop clear business strategy; identify initial partners (MOUs?); build team

Tech: demonstrate novel fermentation process at lab scale, build and validate TEA, generate IP, scale and provide initial product specs

Funding: make pitch; build funding strategy (dilutive/ non-dilutive); raise first round

Tymoshenko

Sacramento, California

Mammoth Biosciences Inc., ex-Zymergen, PhD and post-doc from EPFL and University of Geneva (Switzerland)

With my co-founder, we want to turbo-charge cell-free production by providing abundant and cost-effective energy substrates. Our success will be a paradigm shift in making biomolecules. Our energy substrates will enable cell-free systems to compete with inherently less efficient fermentation. Our innovation is the key to cost-effective enzymatic production of DNA, RNA, proteins, other high-value biomolecules, and biopolymers without growing biomass.

Through iterations of literature and patent searches as well as talking to industry experts, we zeroed in on our initial product and an enzymatic path to producing it. We performed techno-economic analysis to estimate COGS at different scales of production and key variables that affect the COGS. We have planned the initial proof-of-concept experiment to assess known unknowns, and potentially discover unknown unknowns.

In the next few months, we plan to execute lean PoC experiments to assess known unknowns and, possibly, discover some unknown unknowns. Subject to the availability of additional funds we will expedite the progress towards establishing a minimal viable process to make grams of the product and validate our downstream processing approach. We then will share the product samples with select early adopters who can readily use it in their cell-free systems and share their initial feedback on our product's utility. Based on the feedback we'll iterate on our production and downstream processing to 1) improve production yield; 2) scale up production volume; and 3) improve product compatibility with the customer processes.

Sawaya

San Francisco

Azulene Labs (previous: Intel Labs; Harvard @Aspuru-Guzik group)

I want to increase the accuracy of chemical/materials property prediction by 10-50x. I have strong theoretical reasons for believing this is possible. This would greatly reduce R&D costs, as it would allow us to entirely replace many laboratory experiments.

I have founded the company, developed most of the theory, and written some of the code. I have done some consulting work through my startup. I've applied to an ARPA-E grant jointly with another startup.

My big priorities are:

- get to certain MVP milestones

- raise an Angel round (already started)

- purchase significant amounts of computer hardware

- make my first hire

Parikshak

New York City Area

Regeneron, UCSF, UCLA

Generate and leverage large omics data and apply machine learning to identify new therapeutics in neuroscience.

Evaluated competitive landscape, established some proof of concept with existing data from prior work, explored therapeutic indications and feasibility of future clinical trials.

Establish a strategy that maximizes probability of near term successes while setting a foundation for future discovery and differentiation from competition. Delineate experiments, analyses, IP strategy, and necessary funding for next 1-2 years in collaboration with co-founder.

van de Lindt

Bend, OR

NuScale Power, Elementl

I want to reshape heavy polluting industries through the use of high capacity renewable energy sources to reduce emissions. My goal is to build an energy abundant and secure future while not sacrificing efficiency or the environment.

To achieve this goal I have worked in the energy industry specifically Nuclear. Additionally I have begun techno economic analysis of a nuclear design.

Continue an optimization tool in Python that will allow for the optimal reactor design to be constrained on economic variables. Additionally, I will look into conflicts with my current contract.

Gisselberg

Foster City, CA

Currently working on the startup, but previously at Google X and Stanford.

My cofounder and I believe we can fundamentally change how biochemicals are made by unlocking cell-free production. We are developing a unique way to produce energy substrates which is potentially orders of magnitude cheaper then current market rates. This is the missing key needed for synthesizing biomolecules at scale without needing to rely on cells.

We have talked to people in the industry and have completed a techno-economic analysis to make sure there is a solid business strategy behind our idea. We’re at the stage where we’ve planned out initial POC experiments and excited to get into the lab!

Our next major step is to get into the lab and complete our POC experiments to identify and solve any issues with our plan. Our major blockers are access to lab space and equipment. With additional capital we plan on expanding our POC work to make a minimal viable product that we can start distributing to early partners for feedback.

Tay

San Francisco

UCSF

I develop molecular technologies that enable high throughput studies of the human immune system. I seek to combine these technologies and computational approaches to greatly expand our understanding of the immune system and create novel immunotherapies and cures for diseases like cancer and autoimmunity.

I've developed two technologies--one that enables tracking of cell interactions and another that allows high throughput discovery of antigen and antigen receptors. Much effort has been put into validating and scaling these technologies to the levels needed to achieve the applications I envision.

My plan is to continue optimizing and scaling these technologies and in parallel, start to execute various proof of concepts in potential target areas using the platforms to evaluate feasibility.

Bult

Fremont, CA

Currently work at Tesla in Cell Manufacturing. I previously worked at Corning Glass, Dow Chemical, National Renewable Energy Laboratory, and NASA MSFC.

To build a lasting company that can build hard tech to improve the energy landscape in North America. The battery industry is dominated by the APAC region and for North American companies to succeed they require disruptive approaches that will work through scaling. I intend to provide that technology and approach.

I have build process prototypes and flow sheets for the technoeconomic analysis of preferred battery manufacturing design. Over 10 design iterations on process that did not yield acceptable manufacturability have been shelved.

The next step is to identify a suitable industrial location to work on design iterations and to accelerate creation of mock demonstrators, process flow video/fly throughs, intellectual property, and working prototypes.

Roquero Gimenez

San Francisco

Just quit my job at Adela Bio to focus on building.

I hope that multi-cancer early detection becomes one day reality. As of now, the diagnostics industry has failed. I am approaching this from a first-principles approach, working backwards from what a final product should look like. This means using ML at the right places in the product development, and using PCR as the lab workhorse. I am building a computational pipeline to prototype this idea in the most lean way.

I have an end-to-end compbio pipeline that searches for highly specific sequences in a reference-free context, using biological insights on early cancer development to prevent overfitting as much as possible. I have run several experiments leading to candidate cancer-specific sequences, though there's still a long way to go to confirm these findings in-silico before moving to actual PCR.

I think there is enormous amounts of publicly available data to corroborate multiple times over whether a particular sequence is indeed cancer specific. Also, I want those sequences to have documented cancer-related properties, to be shared across multiple species, and harbor some specific patterns that are created through a clear biological process (like alternative splicing). I want to avoid the trap of moving directly into wet-lab testing without having sufficiently de-risked in silico. I think that in a couple months I will know whether I have a robust candidate that makes sense to test in a lab, or whether this whole approach is just not working.

Kim

Boston, MA

Current: Mass General Brigham (Previous: MIT)

I currently work on genomics-related computational/statistical methods for the microbiome. I think that the microbiome is a large missing link in our understanding of health. I want to build algorithms (with or without AI) to understand it better and design non-invasive treatments for some of humankind's persistent ailments.

At MGB we've been working towards a general large-scale understanding of the microbiome -- meaning trying to understand the system as a whole, and working towards predicting emergent behavior that is only possible with an entire microbiome instead of a small, selective culture. For a while now, my working partner Utkarsh Sharma and I have been discussing the untapped potential of combining both genomic and "non-genomic" studies, which we understand as being the missing ingredient in our analysis pipeline.

We want to implement a model that we've dubbed "microbial general intelligence". Even if the end-to-end pipeline (using this model, to translate a microbiome sample into a disease-curing pill) requires more R&D (e.g. clinical trials), the model itself will be of significant interest due to its ability to predict outcomes of microbe-oriented treatments for selective diseases, such as FMT for C.diff.

Sharma

Cambridge, MA

Postdoc at Mass General Brigham (Brigham and Women's Hospital))

One (bad) way to program a computer is by changing the melting point of copper wires or doping of silicon transistors. In contrast, high level intervention via programming languages offers robust, precision control and prevents unintended side-effects. Medicine needs to move in the same direction-- precise, high-level interventions that do not bypass the checks and balances of the human system.

The (gut-)microbiome holds this potential, and consequently for the first time we see the promise of a true cure for many chronic ailments. We are building Microbial General Intelligence (MGI), an AI that will understand the microbiome with context, allowing the development of microbiome therapeutics.

We have shown that despite its immense complexity, the microbiome can indeed be predicted and controlled. In our academic work, we have built state-of-the-art models that predict the future behavior of the gut-microbiome. We have also invented a technology that profiles the impact of the microbiome on the host cheaply and noninvasively [patent filing in progress].

Over the next few months, we want to train our AI models to predict microbiome dynamics in a few-shot manner. We have made blueprints for a few potential custom architectures by gluing together LLMs with our own pieces, and we want to train them on the large amounts of publicly available data that traditional microbiologists cannot leverage. This will help us 1. precisely understand the scaling of our models, which will refine our timeline for our first therapeutic and 2. understand how to make our models generalizable in order to apply to other de-risking industries like animal husbandry and over-the-counter personalized supplements.

Further, we want to explore the market for a Microbial AGI in drug-discovery. Our expectation is that since our AI will be able to understand microbial interactions from genetic information, it will form a cheap and scalable precursor to wet-lab experiments for therapeutic design.

Krauth

Palo Alto, CA

Currently a postdoc at Stanford University in a microfluidics lab, previously I did my PhD in machine learning at UC Berkeley.

A key step in the assay is tagging proteins using a technique called cDNA display. I’m currently working on increasing the yield of this step so more proteins end up getting tagged during the reaction. I’m also scaling up the size of my protein library for a large-scale proof of concept run.

Moore

Costa Mesa, CA

Currently at Anduril Industries, and have been for a few years. Before that I was at an EV startup, and doing prototype engineering for Fiat Chrysler before that.

My work thus far has been disruptive to the US Military Industrial Complex, but the moonshot I see is to build opportunity to build an American Tech Company that democratizes rugged, cutting edge hardware.

I’ve benefited and developed a network of people who believe in me and can support me, as well as a catalogue problems I know I can solve. These problems include rugged edge compute, networking, radio capability, as well as other simpler and more complex products. I’ve also put my heart into 3+ years of experience at Anduril, bringing 13 products to fruition in that time frame.

I’m planning on developing a market map for my ideas, finding two to three more cofounders in addition to the one I have found already. From there we will need to develop a pitch deck, do a hackathon together to understand team chemistry, and decide what our first steps as a company will be. That’s just scratching the surface, but if I sound like I know what I’m doing, I don’t.

Powers

Palo Alto

Stanford (I just finished PhD from Stanford, I still do research there)

My vision is to use the 3D dynamic structures of proteins to rationally design safer, smarter drugs. Most drugs work by binding to proteins like a key fitting into a lock. Thanks to incredible advances just in the last few years, we can now rapidly determine the 3D structures of almost any protein. But these structures are still just static snapshots of very complex, dynamic systems; my work goes a step further, leveraging protein flexibility and motion to uncover hidden opportunities for drug design. For example, we might discover a drug binding pocket that only appears when a protein changes shape in response to particular cellular signals. To do this, I’ve been developing new computational tools, that combine atomic physics simulations and AI. By targeting more specific protein states and discovering concealed drug binding pockets, we can unlock best-in-class medicines that are more precise and effective, with fewer side effects even for well-studied targets. One of my current areas of focus is to create less addictive pain medications with minimal side effects for long-term chronic pain treatment, which could transform patient care for a very common condition.

(1) Speak to more medical/pharma experts to assess how promising this cannabinoid drug is for patients. Is it truly solving an unmet medical need? Obviously, I presently believe it will, but I’m not a medical doctor and might not be seeing all the factors.

(2) Reach out to Pharma and academic labs, to see if they would sign up for partnerships to test/apply our new drug design technology. Is this a widely applicable technology that I could get contracts for right now?

(3) Simultaneously with above, get the 2 papers published so scientific evidence is concrete and public – basically free advertising and credibility. This could realistically be done in 1-2 months.

(4) Finally, present a convincing plan to stakeholders for licensing / optioning one or both of these inventions before anyone else does.

Vasquez Bolanos

San Francisco

Loyal longevity startup, formerly Cornell, UCSD bioengineering

I want to build an intervention that slows down reproductive organ aging and extends our reproductive window. Givings us more time to decide on whether to build a family, build a company or anything else our heart’s desire as lifespans continue to lengthen. I believe there are a few different paths to this goal, some longer than others, I would like to optimize for the quickest path to approval. This may look like leveraging accelerated regulatory pathways, different primary/secondary endpoints, and/or repurposing existing drug and safety profiles.

High level, my hypothesis for aging is that it is cumulative and the earlier you intervene the better, while down the line I believe reversal may be feasible, to generate a positive impact on human longevity now, we should investigate timing with repurposed drugs. My focus has zoned in on reproductive longevity, I have been investigating various repurposed drugs and the intersection of these with existing literature in the area of reproductive organ biology. However I am not sold that oral administration is the best route largely due to timing. Focusing on the decline of the ovarian reserve in women, I believe to have a meaningful impact on maintaining fertility is to intervene earlier closer to the age of 20. As you can imagine, a systemic intervention could prove troublesome, with the brain fully developing by the age of 25 and bone density setting around the age of 30, earlier interventions could possibly risk disrupting the body’s continued maturation. This is where I think localized administration, intravaginaly could prove key for women. Contraception is a great example with far fewer side effects with intravaginal or intrauterine administration. 

On the note of contraception, something else that has crossed my mind, is combing this intervention with contraception. Rational being that, in theory, one would think that contraception should delay ovarian reserve decrease because it is preventing women from ovulating, however, what we have learned from studying contraception in women, is that this is not true. Contraception prevents the release of eggs, but does not prevent the recruitment of primordial follicles and the maturation and atresia of those follicles. Then if you paired contraception with a drug that is capable preventing the recruitment of primordial follicles and maturation of those follicles then the ovarian reserve should be able to be maintained and not experience the same level of decrease as we age.

I have been able to talk to other researchers in the space to see how they’re thinking about this space and what technology is at the frontier (specifically in the ovarian and testes exosome space, which is very high potential but high technical complexity.)

Will admit the following is a bit of stream of consciousness to jot down what I am thinking of.

I need to figure out what is the best way to test the local administration, is it an organoid model of the ovary or a mouse/rat in-vivo study to screen the top few candidates.
I’d like to take advantage of the 5052b FDA pathway of repurposing drugs, but given the different route of administration there likely are additional preclinical safety study needed (ie. Rabbit vagina sensitivity/irritation is the gold standard to the FDA) at what point do I screen this in case it’s a kill?
My intervention has to be exceptionally safe, ensuring it is not hurting women’s fertility, my guess is chronic use and pupping studies in mice or equiv. May be needed. Or along this line, how long of a study is needed at bare min to observe a positive impact on ovarian reserve? Based on the rapamycin menopause study, within 12 weeks there was a positive impact, could likely use similar primary and secondary endpoints. This is part of the reason I think it would be interesting to pair with population of contraceptive users, because they do not want to be pregnant but they would likely not be opposed to taking something that protects their fertility, which allows the risk of having a child on X intervention to be mitigated.

El Nesr

Bay Area, CA

Stanford University

I’m a rising fourth year biophysics PhD Candidate at Stanford University working at the intersection of deep learning x enzyme design, with a broader interest in biological solutions for climate change and human sustainability. My research at the moment involves developing methods for designing new-to-nature enzymes for functions and reactions that currently do not exist; and if they do exist, making them more efficient.

Right now, I have built a lot of the methods involved for developing these biologics: building ML methods to predict protein dynamics, developing algorithms for metal binding, and experimental tests for metal binding. What's left? Putting them together for a proof-of-concept enzyme!

Right now, I am completing a series of experiments to test whether my models do indeed bind the metal they are designed for. Per my PhD, I am doing Zinc. Next, I need to move the model to incorporate other metals including but not limited to Copper, Cobalt, Nickel, and Lead. Next would be testing those proteins out. Over the last two weeks, I have figured out a pipeline and assay to easily test for their binding. For building an enzyme, I am finishing model training for prediction of dynamics and with my collaborator (and maybe future co-founder?) we will test a proof of concept protein there. Over the next couple months, I will need to pick an enzyme to design and merge the two models together. Then I would experimentally test it!

Asmare

Atlanta, GA

Biomedical Microsystems Lab @ Georgia Tech

I want to transform therapeutic cell manufacturing testing from a labor-intensive, fragmented science into an autonomous, massively scalable manufacturing workflow

My co-founder and I have raised almost $500k to de-risk the technology, build a POC device and launch customer discovery campaigns. We've secured agreements to run validation studies with 2 academic labs. Commercial partnerships will be initiated once validation studies are completed favorably.

- Run platform with engineered cells from academic partners

- Continue running customer discovery to identify their pain points in manufacturing and beyond

- Create low volume production run, qualify platforms in preparation for deployments with commercial partners

Wilson

Boston

MIT

Solutions to water scarcity

Developed TRL 5 devices for water production from air

Find out if my tech is worth commercially pursuing.