Blog
One article per project: what it does, how it is built and the technical choices that make it reliable.
Insight Run, generating a training plan without paying for the LLM twice
Insight Run generates training plans with an LLM. Generation takes up to two minutes, the app can be suspended at any time, and an interrupted call is still billed. It runs in Cloudflare Workflows, in four-week blocks saved one by one.
Insight Run, one model per request type
Insight Run calls language models for tasks of very different sizes, from a few sentences of advice to a 16,000-token training plan block. The Cloudflare Worker maps each route to a request type and picks the model from a configuration stored in KV, which can change without shipping the app.