2 min read
Chapter Recovery and a Deterministic Density Study
Adding recovery and validation to the book processor, then running a structured density experiment on the output.
PythonNLPResearchCLIData
Read more
2 min read
Starting Fiction Codec: Multi-Format Book Ingestion
A new project for processing books across formats, with an ingest pipeline, structured storage, and handoff docs.
PythonText ProcessingNLPCLIArchitecture
Read more
2 min read
Adding a QLoRA Training Script to the Benchmark
Adding fine-tuning capability, consolidating docs, and refactoring the training factory.
AIQLoRAFine-TuningPythonDocumentation
Read more
1 min read
Restructuring the Small LLM Benchmark
A single commit reorganizing the benchmark project and adding the features the first pass revealed.
AIBenchmarkingPythonProject Structure
Read more
1 min read
Consolidating the Runtime Core and Parquet Research Tooling
Finishing the dex-trader runtime consolidation and unifying edge-lab's parquet tooling.
RefactorPythonArchitectureTradingParquet
Read more
2 min read
Regime Feeds, Volume Workers, and Curation in Candle Service
Adding a Binance regime feed, backfilling volume, and dropping a concept that didn't earn its complexity.
PythonMarket DataSQLitePerformanceRefactor
Read more
2 min read
Candle Service Fixes: Inclusive Timestamps, User Agents, and Discovery
A day of upstream-integration fixes, plus a new drag-curve worker and a broader discovery query set.
PythonFastAPIMarket DataDebugging
Read more
2 min read
Standing Up Candle Service: A Market Data Aggregation API
From design doc to a deployed FastAPI service with multiple upstream clients, workers, and rate limiting in one evening.
PythonFastAPISQLiteMarket DataDocker
Read more
2 min read
RunSpec, Bootstrap, Bayesian Scoring, and the Pyramid
Consolidating config into a RunSpec, adding bootstrap and Bayesian layers, and turning a research pyramid into a funnel.
PythonArchitectureBayesianBacktestingResearch
Read more
2 min read
134 Commits: Sprints, Ensembles, and a Full Research Pipeline
The largest single day in the repo—a three-track roadmap executed as six sprints, ending with walk-forward ensembles and a filter-quality bundle.
PythonBacktestingMachine LearningArchitectureResearch
Read more
2 min read
Building Edge Lab and Porting It to a Triple-Arc XPU Rig
Phase 1 through 9 of a trading research platform, then a hard pivot from numba-dpex to torch.xpu, Triton, and IPEX.
PythonTritonIntel ArcPerformanceBacktesting
Read more
2 min read
Making the Venue Churn Pipeline Deterministic
A late night stabilizing a fleet of model profiles, capping inference hangs, and promoting a faster quantized fleet.
PythonLocal LLMReliabilityAIInfrastructure
Read more