All Operator Stories Watch on YouTube
Operator Stories1:08:01
Who manages the work when AI writes the code? | Corey Pritchard
Corey Pritchard joins Dave to explain why he built alterac.ai to coordinate coding agents. They discuss planning, context, dependencies, fresh-context reviews and the engineering judgement needed to verify AI-generated work.
Hosts
Guest
- Corey Pritchardalterac.ai
Show notes
Operator story
Corey Pritchard joins Dave Pengelley to discuss alterac.ai, the coordination layer he built around local coding agents. He describes the engineer's role in planning work, supplying context and reviewing results, and how he uses the product to develop the product itself.
In the conversation
- Managing plans, tasks and dependencies across coding agents.
- Passing context from a task into an agent's working session.
- Using Git worktrees and choosing when parallel work makes sense.
- Reviewing results with fresh context and handing work to another agent.
- Applying engineering judgement to verification and security.
- Personal AI agents, outreach and useful work overnight.
Product behaviour and workflow examples are Corey's account at the time of the interview, not independent benchmarks, security guarantees or promises that roadmap features are available.
Transcript
Transcript coming soon.