Guardrails, not gates: Scaling juniors in the AI era
summary
Claim: scaling juniors succeeds by building guardrails not gates — concrete tools and metrics. Wise deploys LLM-powered tooling (Codex, Claude Code) into a Wise Agentic Platform and Wise Brain MCP, ships three Wise Star skills (Wise Star Breakdown, Wise Star Code Flow, beta Review with me), runs CI agents and a PR review bot that saves ~2 hours per review. Onboarding to ten PRs halves for agentic grads, 50 hires a year among ~1,000 engineers, PR throughput and change‑fail rate show net improvement.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:30Early career challenges
- 3:41Wise Star program
- 6:54AI adoption challenges
- 9:42Impact metrics
- 11:51Wise Agentic Platform
- 14:15PR review bot
- 16:18Speed versus security
- 18:43Senior perspectives
- 20:47Key takeaways
related talks
08mBuilding an AI Platform Your Engineers Actually Trust
34mBuilding Guardrails for AI Coding Systems
51mMake Quality Executable: Verification Gates for Agentic Engineering at Scale
30mAgentic development panel
31mDefying Gravity: Teaching AI to Write Better Ruby
01h 01m"Looks Good to Me": A Practical Guide to Handling AI-Generated Code
42m10x Engineering: How GenAI Tools Transformed My Team's Productivity
26mBeyond the code: Leading engineering teams in the GenAI era