AI coding tools generate plausible but wrong SQL constantly. The fix isn't waiting for a smarter model.

AI skills are markdown files that encode domain knowledge into coding tools. No framework, just structured text in a repo.

The loop: code → review → handoff → skills update. Every session makes the next one smarter. One aggregation bug became a permanent rule enforced automatically.

Dori Wilson broke it all down at Data Debug SF. Full writeup:
https://blog.reccehq.com/ai-skills-for-analytics-eng

#DataEngineering #AI

Our own Kent Chen wrote up the multi-agent architecture the team built for Recce's AI Data Review.

Single agent kept forgetting findings as PRs got complex. Fix: orchestrator + two specialists, each with its own 200k context window.

One subagent fetches full PR context via a single GitHub GraphQL MCP call (replaced 5-10 gh CLI round-trips). The other explores data through 6 Recce MCP tools: lineage_diff, schema_diff, row_count_diff, custom queries.

"Pandas in 2011 was essentially book-driven development, quite literally." Wes wrote features because he needed them for the book chapter.

#pandasPython #DataRenegades

"If I had taken three years longer to do things the right way, it would have been too late." -- Wes McKinney on why pandas shipped imperfect and won.

#pandasPython #DataRenegades

"This is bad. Why haven't you fixed this yet? I would have already fixed this today with Claude code." -- Wes McKinney on radical accountability for software vendors.

#AIcodingagent #DataRenegades

Starting soon: Bauplan x Recce live session on safe AI automation from branch to production. https://luma.com/mm3gsalo?tk=S6sKK7

#DataEngineering #AI

"I would just wake up and write Python code." Wes McKinney on the founder hours that created pandas.

#DataRenegades #pandasPython

Scott Breitenother jumped into every Slack thread at Brooklyn Data. It made the company fast and him the bottleneck. His fix: subscribe to replies, don't comment, check back in 3 hours.

#Leadership #DataTeams