Kinetic Painter: getting latency out of the way
A conversation with Jev has me thinking about returning to Kinetic Painter.
Lynn Cole / Laboratory
Laboratory
Experiments, research notes, mad science, and things in progress.
Can play-training teach engineering taste?
A small Qwen training run tests whether play, critique and revision can change engineering taste while preserving correctness. The results guide the next replication run.
Building the play-training dataset
Building a play-training dataset takes longer than the neat plan suggested. These are the challenge counts, solution trajectories and training signals so far.
Hubnot: issue revisions, retries, and descriptive authority
Hubnot adds issue structure, explicit revision history, retry handling and stale-update checks while keeping assignment separate from execution authority.
Nano-Kitty: long-horizon work beyond a single session
Nano-Kitty combines Spec Kitty’s planning with a long-running agent harness, aiming to carry complex work beyond any single session.
Nano-Kitty: isolated worktrees and a heterogeneous litter
Nano-Kitty isolates parallel worktrees and processes while building toward a model-agnostic, long-horizon agent system.
Planning a self-play experiment for a coding model
A planned self-play experiment uses compiler verification and multiple reviewers to turn a small Qwen model into a stronger Go developer.
Video workflows built around deliverables
A video-generation experiment organizes work around deliverables rather than isolated renders, while optimization exposes the shape of the model pipeline.
Glamin in a local inference stack
Putting Glamin into a local inference stack turns vector-space state changes into a possible interface for tools, memory and small applications.
Agents A1 and Qwen: coding-workflow observations
A coding-workflow comparison of Agents A1 and Qwen finds different failure modes across Go, Zig and Next.js, and a useful response to QA feedback.
Finding Qwen’s limit on a vague Zig assignment
A deliberately vague Zig assignment pushes Qwen through planning, implementation and the complexity wall. The failure is part of the evaluation.
Video restoration before the information is gone
AI upscaling can improve archival footage while making clean video look waxy. The next experiment deliberately degrades the source before restoration.