Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Tonk: Building Malleable LLM Software
Explore how Tonk enables non‑coders to quickly update multiplayer applets by limiting context to the frontend and using extensive hint‑file prompting.
We’ve been developing Tonk for shareable, multiplayer applets enriched with context and intelligent from across your life. What makes developing these applets a little different is that they are “malleable”, that is, they are meant to be updated quickly by the users of the application through vibecoding. At first, we struggled to keep the coding agents from going off the rails, but now our framework, to our surprise, is totally usable by non-coders.
I’ll talk about the two ways we achieve this:
1) eliminating the need for context beyond the frontend (ie. what the agent can see)
2) prompting with lots and lots of hint files
I can also talk about a few failed experiments: using a recursive task definition framework for the agent and injecting context through MCP plugins.
Tonk: TypeScript toolkit for local-first, real-time data applications.
Related projects
Keeping an "AI" on LLMs with Langfuse
London
Learn how to self‑host Langfuse for LLM observability, covering setup, tracking user queries, inputs, retrieved data, and practical…
LLM.f90 - Minimal Large Language Model Inference Framework
Toronto
A low‑dependency Fortran framework for LLM inference, showing zero‑dependency implementation, matrix operations, and support for Llama, Phi, and…
Thinking LLMs
Los Angeles
This talk explains how to generate synthetic data for training custom o1 style language models using methods from…
AI on Trial: Fine-Tuning LLMs to Judge other LLMs; data, results and challenges
London
We will present data, experiments, results, and challenges of training language models to evaluate other models, plus insights…
Using LLMs to automate content moderation
Dublin
A practical overview of using retrieval‑augmented generation, fine‑tuning, and prompt engineering for content moderation, focusing on accuracy, consistency,…
LLM Evaluations in Practice
Amsterdam
Learn about a practical setup for LLM evaluation in production, sharing hard-earned lessons for guiding prompt and code…