Share
Explore

icon picker
Getting started with Coda AI

New to Coda?

Check out
our guides
⁠
to learning Coda and
sign up
⁠
for a “fundamentals of Coda” webinar.

Learn by....


Watching

⁠
Sign up
⁠
for our webinar on “how to add Coda AI to your workflows”
Coda-led demos and workshops also available for enterprise customers.

⁠

Tips for implementation

Share an internal memo: Remind employees of your AI policies, the approved list of AI tools, and how to access them. This will help employees feel comfortable embracing them, and underscore which ones they should not use.
Establish an AI committee: Creating a cross-functional AI committee or a “tiger team” can leverage the expertise and enthusiasm of early adopters. Many might see this as a way to gain visibility across their org, or to learn a new skill.
Training programs: Coda-led or company-led training programs can help employees learn best practices for their usage. These programs can be supplemented by training sessions within each team, potentially led by a member of the AI committee. Contact your account team to set up a Coda AI training session.
One-off support: Create a dedicated Slack or Teams channel for AI-related questions to foster a collaborative learning environment. Encourage others to share all wins (including small ones) as well as “fails” so others can give suggestions and help troubleshoot. This will also give you great visibility into how Coda AI is being used!
Encourage cross-functional collaboration: Offsites and hackathons provide a creative environment, and often bring together teams that don’t typically collaborate. Dedicate time to exploring AI applications, so co-workers can foster new ideas, and spread best practices.
Embrace different adoption rates: Not everyone will be an early adopter of AI—encourage curiosity and provide support for different learning paces. Ask managers to keep this in mind as they mentor their team and new employees.

Tips for writing prompts

Test early and often. Be patient and diagnose output to help you learn.
The earlier words are in the prompt the more importance the AI gives to them.
The shorter the prompt the better BUT that doesn’t mean it doesn’t need to be longer to work.
Try the simplest version of what you’d like it to do first, test that then add more guidelines and examples to correct it.
1. Provide context
Think about AI as a robot coming to earth and learning everything for the first time. Be blunt and specific, telling it exactly what you want to do.
Tell AI how the content will be used, such as a blog post or email, and define your target audience if possible.
Indicate tone, style or other nuances.
2. Give it a persona
Tell AI to write in the specific style of someone, or based on a given role.
3. Give examples
Provide 1-2 starter suggestions.
You can write both “good” and “bad” examples.
4. Define format
Indicate length, e.g. 50 words or less.
Give example of preferred output (bullets, sentence, H2 header).
Want to print your doc?
This is not the way.
Try clicking the ⋯ next to your doc name or using a keyboard shortcut (
CtrlP
) instead.