Wekinator

Aiden is Weco's autonomous multi-agent machine-learning research system. It retrieves research papers and GitHub implementations, proposes and audits ideas, runs training experiments, retains code changes that improve a language model's evaluation score, and submits pull requests for promising results in the Parameter Golf competition. Its research loop repeatedly generates proposals and evaluates them against a fixed metric, with experiment branches and scores made inspectable through Weco's dashboard.

W&B Weave is an AI evaluation and observability tool that stores and displays per-run trajectories behind leaderboard results. In the WolfBench workflow, it preserves inspectable traces of commands, messages, failures, and other run evidence so reviewers can trace a score back to the conditions that produced it.

Weco AI is an autoresearch agent platform for running and inspecting optimization experiments. It repeatedly proposes code changes, evaluates them against a fixed command and measurable goal, keeps changes that improve the score, and discards weaker attempts. Its dashboard presents the search as an experiment tree with scores and diffs, allowing users to branch from promising runs and steer an ongoing search toward a different idea. The workshop materials demonstrate the approach with Mini Parameter Golf and a Flappy Bird controller evaluated across multiple conditions, including hidden validation data and limits on editable code.

WERM (Wiring-Encoded Recurrent Modulator) is a computational-neuroscience simulation that models a 302-neuron Caenorhabditis elegans network using the worm's published connectome. It runs as a rate network with a liquid-time-constant-style gate; the model treats the 28 GABA-releasing neurons as inhibitory and the remaining neurons as excitatory, and supports simulated sensory stimuli such as nose and tail touch. A steering layer maps the simulated worm state to language-model sampling settings and a tone prompt. WERM can run in a browser, in Node.js, or connect on the user's computer to Ollama or another local server implementing the OpenAI chat-completions API. The repository reports that tail and nose stimuli produce the expected forward or backward network responses in its tests, while the effect of the language-model steering has not yet been measured. The project requires Node 20 or newer, has no runtime dependencies, includes scripts for rerunning parameter sweeps and experiments, and is licensed under the MIT license. It is an intentionally small model without muscles, body physics, neuropeptides, learning, or comparison against recordings from living worms.

Weave is a meeting intelligence tool that listens to spoken conversations and generates a live, editable map of the thinking during a discussion (decisions, open questions, and who said what). Maps are saved, replayable, and shareable for review and asynchronous collaboration.