The internet has been going crazy with videos of people building the most basic apps using AI, but for developers working on more complex code, backend development, and building on an existing codebase, it's still quite limited.

Here are 4 use cases for AI in software development apart from code generation.

1) Turn meeting into a release.
There are really good AI transcript tools out there, which integrate with Google Meets, Zoom etc. If you're having an in person meeting you can also open an app on your phone and stick it in the middle of the table. AI can summarize and turn the transcript into technical specifications; the specs into an implementation plan and the implementation plan into a PR. Using Fine you can review and make adjustments at each step of the way.

2) Adding context to issues
When a new issue is created in GitHub, Linear, Sentry etc., have the AI search your codebase for relevant files and similar issues and add it as a comment. When you come to start working on the issue, you'll already have a more in-depth brief to refer to and jump start the process. You can do this using Fine workflows, so it happens automatically.

3) Self-healing CI/CD
Use a platform such as CircleCI to monitor your CI/CD rather than doing it manually. When the pipeline fails have AI analyze the logs, generate a troubleshoot report and suggest or commit a fix depending on your preference.

4) Generate Changelog
No coding involved, no risks of hallucinations - have AI pull all PRs and issues from the day/week and generate a changelog report for your stakeholders. Super straightforward.

Interested to hear other use cases apart from just writing code with GH-CP and Cursor, please share in the comments :)

These ideas are all from a workshop by Dan Leshem who is CTO at Fine. Fine is $15 per month with a free trial.