Coaching for Teams
Level up your team for the AI era, on your real code
AI moved the bottleneck from writing code to judging it: reviewing what the model produces, directing it well, and knowing when to reach past Python. I run hands-on team programs built around deep, human code review, the part AI can't supply and the thing that makes new habits stick. Two tracks, or custom 1:1 for your engineers.
Agentic AI: ship production agents
Your team is already shipping AI features. The gap is the engineering around the model. In six weeks they build and deploy a real agent, hardened under review, and leave able to build the next one.
- Production architecture: function calling, structured outputs, service layers, tests, CI/CD
- The review layer most courses skip: catching what AI gets wrong, one standard across the team
- A deployed agent per developer, not slides
Rust: remove Python bottlenecks
Keep Python where it shines and move the hot path (CPU, latency, memory) into safe, fast Rust via PyO3, the same bridge behind Pydantic and Polars. Your team also gains the judgment to know when Rust earns its place.
- A real component shipped: a parser bridged to Python with PyO3, benchmarked against CPython
- Ownership, lifetimes, error handling and concurrency, taught by building and reviewed on every PR
- The memory-safety case to back the decision to stakeholders
Is your team ready for AI-era engineering?
Two things stall teams right now: code review that can't keep pace with AI output, and performance you can't push past in Python. Take the 30-second check, then let's talk through what it surfaces.
Check what's true for your team:
Who runs it
Coached by Bob Belderbos: 20 years building software, co-founder of Pybites, more than 150 developers coached. Rust and Agentic AI team sprints are co-led with specialist engineers Juanjo Expósito and Jim Hodapp, so there's delivery depth behind an engagement, not a single coach.
The first step is a conversation
Tell me about your team and your code, and we'll work out the right track (a cohort, team seats, or custom 1:1).