Let's covering the basics, like setting permissions and sandboxing, writing the AGENTS.md file, setting up rules & the project’s source of truth, and digging into how memory works. But also how to build and run subagents & Skills, how to combine them, and how to evaluate them.
We’re building a tool for kids to create AI videos but when users create clips with different models, or in general stitch clips together, the voices and sound will drift. I'm going through building a cheap pipeline that creates consistent voices.
I scouted 93 docs that could act as a proxy for what companies use OCR for — handwritten notes, tables, financial legacy docs— and tested them on 14 different OCR engines such as smaller specialized open-source vision models, document parsing tools and general frontier models. Let me show you which won for easy, medium and the hard tiers and how to save money when building OCR pipelines.
Let's talk about how to decrease agent costs by 90% using various techniques. We'll go through prompt caching works and why it’s a quick win, semantic caching, lazy-loading tools and MCPs, routing and cascading, delegating to subagents, and a bit on keeping the context clean.
