Domain-specific coaching agent / SELECTED ENGINEERING WORK
Snooker Coach AI
Level-aware snooker coaching for rules, technique, drills, practice plans, strategy, and match review.
This overview describes my engineering work on this project, including its purpose, implementation, and technical focus. Project illustrations are abstract; private assets and internal implementation details are not published here.
Overview
Level-aware snooker coaching for rules, technique, drills, practice plans, strategy, and match review. My work connects the user experience with the data, application logic, and integrations behind these workflows.
Problem addressed
A useful coaching agent needs guidance appropriate to the player and clear boundaries around its domain.
Solution
Built a Next.js and TypeScript coaching agent with OpenAI/Ollama provider abstraction, typed drill content, and deterministic guards before model calls.
Key features
- Level-aware coaching
- Rules and technique guidance
- Cue-ball control drills
- Practice plans and strategy
- Match-note review
- Domain and safety guards
- Unit tests and prompt evaluations
Technical outcome
A focused coaching agent with interchangeable model providers, deterministic domain boundaries, and testable coaching content.
Interested in the engineering behind this work? Get in touch to discuss the approach and a similar product.