The notebook, page by page
The archive.
Experiments, implementation notes, and the details behind a decision. Sorted by the most recent publication or revision.
30 articles
- ↗
PostgreSQL Full-Text Search: Test the Search You Actually Need
Test PostgreSQL full-text search with a disposable database: phrases, field weights, zero-price filters, updates and the limits of a tiny fixture.
Includes verification notes - ↗
FAISS, Pinecone and Weaviate: Compare the Search Contract First
Compare FAISS, Pinecone and Weaviate search contracts through a runnable FAISS cosine, filtering and persistence experiment.
Includes verification notes - ↗
Inside Elasticsearch: Terms, Positions and the Search You Asked For
Trace terms, positions and BM25 with a tested Python model, then separate those mechanics from Elasticsearch analysis, refresh and shard behavior.
Includes verification notes - ↗
Build RAG From Scratch in Python: Test Retrieval With LangChain
Build a tested LangChain retrieval-to-prompt pipeline with source IDs, access scopes and repeatable indexing, before adding model generation.
Includes verification notes - ↗
Your first Ruby gem: test the package, not just the source file
Build a small slug gem with a precise Unicode policy, Minitest checks and an isolated build-install-load test. No publishing account required.
Includes verification notes - ↗
ngrok With Local LLMs: Test Authentication Before Sharing Ollama
Test the HTTP authentication boundary before sharing Ollama or LM Studio through ngrok, with a runnable local preflight and explicit integration limits.
Includes verification notes - ↗
Ruby design patterns: make undo and redo survive a new branch
Implement the Command pattern with a tested history cursor. Handle undo-all, branching, failed commands and external edits without negative-index surprises.
Includes verification notes - ↗
Ruby metaprogramming: generate a small API without hiding its rules
Build a typed settings DSL with define_method, then test reflection, validation, inheritance and shared defaults. No eval or method_missing required.
Includes verification notes - ↗
Sinatra, Rails and Roda: Compare One HTTP Contract
Compare Sinatra, Roda and Rails API controller routing against one JSON request contract, with 33 passing Rack tests and locked Ruby dependencies.
Includes verification notes - ↗
Predicting Exam Outcomes in Ruby: Split Before You Scale
Use Rumale to fit preprocessing on training rows only, evaluate a synthetic exam-outcome fixture and inspect threshold errors against a baseline.
Includes verification notes - ↗
Implementing the Rasch Model in Python: Fix the Scale Before Fitting
Build and test a penalized Rasch estimator in Python, with a zero-mean difficulty constraint, checked gradients and extreme response cases.
Includes verification notes - ↗
Ruby functional pipelines: make failure a value that stops the next step
Implement a small Ok/Err pipeline in Ruby. Test short-circuiting, map versus bind, false and nil values, and errors that should still raise.
Includes verification notes - ↗
Grok-1's Mixture of Experts: Active Parameters Are Not Stored Parameters
Understand Grok-1 sparse routing with an executable eight-expert NumPy experiment that separates active computation from stored weights.
Includes verification notes - ↗
Timing LLM inference in Python without inventing a CUDA speedup
Build a tested inference timer around an offline GPT-2 fixture, count generated tokens correctly and state exactly what a CUDA measurement would include.
Includes verification notes - ↗
LeetCode in Ruby: test the invariant behind three useful patterns
Implement lower-bound search, a unique-character window and a min-heap in Ruby. Check duplicates, empty input and seeded comparisons with simple oracles.
Includes verification notes - ↗
How to Train an LLM: Verify the Training Step Before Scaling It
Run a tiny Hugging Face language model on CPU and test label alignment, padding and saved weights while training and holdout losses diverge.
Includes verification notes - ↗
Ruby to PyTorch: test the process boundary before CUDA
Call a real PyTorch worker from Ruby with an explicit JSON contract, tested error paths and a clear boundary between CPU checks and CUDA execution.
Includes verification notes - ↗
A Zilliqa allowance contract needs a withdrawal clock
Test the missing withdrawal interval in a Zilliqa allowance design with a JavaScript state model covering repeated calls, exact balances and integer limits.
Includes verification notes - ↗
Build a Bayesian text classifier in Ruby, then check the arithmetic
Implement multinomial Naive Bayes with word counts, smoothing and log scores. Test unknown words, empty input and a small held-out fixture.
Includes verification notes - ↗
Amazon Transcribe in Ruby: Separate Starting, Waiting and Reading
Separate Amazon Transcribe submission, bounded polling and authenticated S3 output reading in Ruby, with tested SDK stubs for failures and empty results.
Includes verification notes - ↗
A Ruby Coin Collector: Gosu Rendering, Testable Game Rules
Build a Gosu coin collector with normalized movement and testable collisions: eight Ruby model tests and a short native window smoke run.
Includes verification notes - ↗
Functional programming in Ruby: keep the export, lose the mutation
Build a tested order-to-CSV pipeline with map, select and sum. Check input ownership, quoting, invalid rows and the limits of lazy evaluation.
Includes verification notes - ↗
Ruby vs Elixir: compare the state transition before the syntax
Run the same lost-update experiment in Ruby and Elixir. See why a mutex or single-owner process needs to protect the whole operation.
Includes verification notes - ↗
Medical marketing meets a full booking queue
A tested Python queue model shows how booking capacity changes a medical marketing funnel, and where its synthetic assumptions stop.
Includes verification notes - ↗
Ruby and Sonic Pi: Build a Four-Beat Phrase You Can Inspect
Compose a four-beat Sonic Pi phrase and inspect its note schedule with seven Ruby tests, while keeping score checks separate from actual audio playback.
Includes verification notes - ↗
Speech Recognition in Ruby: Send Audio Bytes, Handle Every Result
Build a Google Speech-to-Text V1 request from local audio bytes and extract segmented results, with real Ruby protobuf and SDK method tests offline.
Includes verification notes - ↗
Machine Learning in Ruby: Train, Check and Reload a Ruby-FANN Network
Train an actual Ruby-FANN XOR network, validate inputs and reload saved weights using a documented Ruby 3.3 compatibility build.
Includes verification notes - ↗
Healthcare ML: a perfect score can be a broken split
A reproducible Python example shows patient leakage turning memorization into perfect accuracy, then checks how prevalence changes positive predictions.
Includes verification notes - ↗
QLoRA on a Single GPU: Does a 7B Model Fit in 8GB?
Use a tested Python memory calculator to separate four-bit weights from QLoRA training memory, and measure the parts an 8GB claim leaves out.
Includes verification notes - ↗
A unique key doesn't make a Sidekiq job retry-safe
A Ruby experiment shows why a unique payment record can hide unfinished work, and how retries can resume it with one stable operation key.
Includes verification notes
No notes match this search. Try another word or clear the topic filter.