Research desk
Research
Experiments and benchmarks, reported the way they should be: hypothesis first, method published, code linked, and the limitations stated rather than buried.
How to read these
Every study states its hypothesis before its result, links its data and notebook, and ends with what it could not establish. Numbers here are a starting point for your own measurement, not a conclusion to adopt.
Market research
Does RSI Actually Work?
Testing the classic RSI(14) mean-reversion rule across 20 years and 500 instruments, with costs included and multiple-testing accounted for.
IntermediateDifficulty: IntermediatePython · Polars · DuckDB
NotebookDataset
Benchmarks
DuckDB vs pandas vs Polars on 50 Million Rows
A reproducible benchmark across six workload shapes, reporting memory as well as time — because memory is what actually stops the job.
BeginnerDifficulty: BeginnerDuckDB · Polars · pandas
Notebook
Evaluation
Six RAG Chunking Strategies Compared
Fixed, recursive, semantic, structural, sentence-window and parent-document chunking measured on retrieval recall across three corpora.
IntermediateDifficulty: IntermediatePython · Ragas · pgvector
Notebook