Oct 23, 2026
SQLite extension that runs SPLADE sparse encoders at insert time for model-free semantic search
The extension loads a llama.cpp GGUF plus extracted MLM head weights, runs the encoder and head over ggml during INSERT to produce posting lists stored inside the database file, then answers MATCH queries with only tokenization and an exact scatter-add over the static weight table. This moves all transformer work to write time, giving embedded and read-heavy applications a drop-in semantic search primitive that avoids both vector databases and query-time model execution.
Launcher for seeding Forgejo instances and orchestrating sandboxed LLM agent turns across interdependent repositories
A coordinator loop wakes one maintainer agent per repository on a fixed cadence, each turn executing branches, pull requests, reviews, and releases inside a Harbor sandbox while a separate evaluator scores downstream dependents the agents never observe. The resulting checkpoint-to-bundle pipeline supports controlled studies of emergent multi-agent software engineering but remains specialized tooling for research groups rather than a general developer platform.
Executable RL environment that lets agents diagnose and repair distributed PyTorch training failures inside sandboxed execution
The environment launches synthetic multi-process Gloo jobs, injects one of four fault families, captures trusted launcher records of batches and checkpoints, then re-executes every candidate patch inside a network-less gVisor sandbox so an external verifier can replay optimizer transitions and sample accounting. This creates a narrow but well-defined benchmark for LLM agents and RL methods that must produce repairs that are both executable and semantically correct rather than merely passing superficial checks.
