Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Context Management
Live demo of an AI framework using local models to manage context for coding agents, illustrating techniques for efficient code generation and tool integration.
I’m the guy in the audience who talked about local models and having my own AI framework for tasks like coding, and we talked about doing a “video” demo. This will be that demo, where I’ll show how I manage context for coding agents.
Related projects
Context Engineering
Dallas Fort Worth
Learn how ReKnew’s Context Engineering integrates Dara into enterprise AI workflows, enabling practical agentic and generative AI applications…
How great context selection leveled up Bismuth
San Francisco
Explore Bismuth, a terminal coding agent. This talk details its advanced context selection, file and symbol discovery, and…
Agent Eval and Scores at the Thread Level
Seattle
Learn how to capture, score, and analyze LLM agent interactions at the thread level, identifying failures and aggregating…
Taskplain: Task management designed for coding agents, living right in your repo.
Seattle
Learn how to manage coding‑agent tasks directly in your repository using plain Markdown files, enabling human oversight, async…
Beads: My coding agent memory bank & planning system
Seattle
Live demo of using Beads to coordinate multiple coding agents across projects, converting markdown plans into a queryable…
The Context Lake: From Low-Code Workflows to an Organizational Memory
Medellín
Learn how to build a centralized Context Lake, then use low‑code Zapier automations or an AWS event‑driven service…