What practice customizes AI coding agents to your workflow?

Sagnik Nandy

Chief Technology Officer · Docusign

Fine-tune them

Sagnik Nandy says his team fine-tunes the agents for common cases, since the same kinds of tasks are assigned repeatedly, so you can fine-tune them to make sure it's done right.

The answer starts at 22:23.

�S

🦄 Peter Steinberger

Founder · Amantus Machina

Design a pipeline

Peter Steinberger recommends designing a pipeline from “this is what I want” to “this is the outcome,” saying the better you do that, the better your outcome will be.

Nan Yu

Head of Product · Linear

Define clear goal

Nan Yu recommends giving the agent a really good defined goal, saying they're smart enough to get what they need if you do.

Mark Papermaster

CTO and EVP · AMD

Invest time to train

Mark Papermaster recommends investing time to tailor and train the model of your choice to you, telling it who you are, what you like and your style, because he says that makes it most productive for you.

Daniel Herrington

Chief AI Officer · The Zebra

Create a skills repo

Daniel Herrington says their most successful practice is a "skills of skills" repo, where a GetSkills MD file pulls all skills files from git so they're consistent across developers.

👨🏾‍💻 Ben Ilegbodu

Senior Software Engineer, TV Client Foundations · Netflix

Create intent markdown

Ben Ilegbodu recommends having AI coding agents create an intent markdown file that captures what the user is trying to do, so the agent can update it or avoid changes that break existing features.