- Consistency over novelty: follow existing patterns; don’t add tools or frameworks unless they already exist in the repo or the user approves.
- Simplicity first: prefer the fewest moving parts that satisfy requirements.
- No assumptions: verify dependencies, versions, and APIs against the codebase or docs; ask when in doubt.
- Transparency: explain the rationale behind changes and architectural decisions.
- Security by default: never expose or persist secrets.
All the interesting stuff about Data Science that i've found.
Most are from Toward Data Science (TDS) blog.
(I have a few hundred blog posts in the queue to post... 😟)