Coursework, revived.

• ·A bring-your-own-key evaluation asks whether an LLM can find the best card grouping, scored by the same rules engine against an exact solver and a greedy baseline, with Wilson intervals.
• ·A coverage simulation showed the bootstrap intervals run narrow on small samples, so the default became 30 hands and those intervals are labelled nominal.

Original

Python 3 · Grok Learning

2026 revival

Shared stack + Web Workers · fast-check · zod · IndexedDB

Skills

• Vote counting (first past the post, second-preference runoff, instant runoff)
• Breadth-first search
• Uniform-cost search
• Exhaustive set-partition search
• Parity testing against Python
• Property-based testing with fast-check
• Wilson and bootstrap intervals
• Interval coverage simulation
• LLM evaluation against an exact solver
• Content Security Policy for BYOK keys

• ·A Rules vs LLM harness tests whether a language model can follow the cleansing procedure, scored against the rule engine with Wilson intervals and paired tests. No LLM results are published.
• ·Bring-your-own-key calls go straight from the browser, are restricted by a Content Security Policy and are recorded in a local audit log.

Original

C99 · gcc

2026 revival

Shared stack + Shiki · fast-check · zod · Anthropic SDK

Skills

• C string processing
• Differential testing against a compiled binary
• Property-based testing
• Wilson and bootstrap intervals
• Exact McNemar test
• Rules vs LLM evaluation design
• BYOK AI audit logging
• Content Security Policy

• ·Re-running the originals surfaced three differences the reports did not mention, such as two score rules and an unseeded k-means, and the lab lets you switch between them.
• ·An optional LLM-as-a-judge harness compares a model with the 2020 matcher on a seeded, stratified sample of candidate pairs, run on the visitor's own key and logged locally.

Original

Python 3 · requests · BeautifulSoup · pandas · NumPy · matplotlib · textdistance · fuzzywuzzy · scikit-learn

2026 revival

Shared stack + Web Workers · zod · Anthropic SDK

Skills

• Breadth-first web crawling
• Regex information extraction
• Record linkage and blocking
• k-NN and decision tree classification
• PCA and feature selection
• Repeated stratified cross-validation with Nadeau-Bengio correction
• Wilson and bootstrap intervals
• Exact McNemar test
• LLM-as-a-judge evaluation
• BYOK AI audit logging

• ·An A/B test designer for the late-discount rule covers sample size, a seeded simulation with a known injected effect, and a peeking warning backed by 10,000 A/A runs.
• ·Adds an append-only audit trail and an optional bring-your-own-key shift summary that never sees personal data.
• ·Production now runs on a shared Turso database, so a customer's order reaches the vendor board and the audit tables persist. Decision records explain the earlier per-instance demo storage and why it was replaced.

Original

Node.js · Express · Handlebars · MongoDB Atlas · Mongoose · Passport · Google Maps JS API · Jest · Heroku

2026 revival

Shared stack + SQLite (libSQL) · Turso · Drizzle ORM · MapLibre GL · SWR · zod · Vercel

Skills

• Full-stack web development
• Session authentication with bcrypt
• Parity testing against the original JavaScript
• Kaplan-Meier survival analysis
• A/B test design and power analysis
• Permutation tests and Newcombe intervals
• Append-only audit trail with database triggers
• BYOK AI with server-side record verification

• ·Zone maps, busiest routes and day-by-day weather, events and collisions for all 263 taxi zones, served from a 15 MB read-only SQLite file holding aggregates only, with no trip-level records.
• ·An optional bring-your-own-key 'Ask the data' page turns plain-English questions into SQL that the visitor reviews before it runs, with every call labelled and logged.

Original

Python · PySpark · Spark MLlib · pandas · Folium · Jupyter

2026 revival

Shared stack + DuckDB · SQLite · MapLibre GL · Recharts · zod · Anthropic SDK

Skills

• Cleaning 84 million rows in documented rounds
• Joining weather, event and collision data
• Elastic-net regression with hand-written 10-fold CV
• Temporal hold-out and residual diagnostics
• Robust standard errors and conformal intervals
• DuckDB pipelines with logged row counts
• Zone maps and route views with MapLibre GL
• BYOK plain-English questions with SQL review

• ·Added a bias-variance decomposition, per-source recovery intervals and paired re-tests of the 2021 claims, with decision records that say where the original reasoning does not hold.
• ·An optional 'Explain this figure' feature runs on the visitor's own API key, checks each number in the answer against the figure's summary and logs every call locally.

Original

Python · Jupyter · NumPy · pandas · scikit-learn · matplotlib · seaborn · R

2026 revival

Shared stack + KaTeX · Web Workers · zod

Skills

• Synthetic spatiotemporal data simulation
• Least squares, ridge and lasso regression
• Principal component regression via SVD
• Monte Carlo tuning of the lasso penalty
• Bias-variance decomposition
• Paired comparisons with common random numbers
• Bootstrap and Student-t intervals
• Port of NumPy's MT19937 RandomState
• BYOK LLM figure explanations with number checks

• ·An optional bring-your-own-key meeting-note assistant removes names, e-mails, phone numbers and addresses in the browser before any call, and a 32-note harness compares its follow-ups with a rule-based baseline note by note (paired bootstrap, exact sign test).
• ·Replaced the 2021 services: MapLibre with OpenFreeMap tiles and Photon geocoding instead of Google Maps, an on-screen demo inbox instead of Gmail, and a shared Turso database instead of MongoDB Atlas.
• ·'Try as guest' creates a private 24-hour sandbox with 25 contacts and 40 meetings. The Insights page shows meetings per week with a seeded bootstrap interval, and every account can export or delete everything stored about it.

Original

React 16 · Material UI · Express · MongoDB · Mongoose · Passport (JWT) · Google Maps API · Jest · Heroku

2026 revival

Shared stack + SQLite (libSQL) · Turso · Drizzle ORM · MapLibre GL · jose · zod · Vercel

Skills

• Scrum delivery
• Mobile-first UI design
• Session authentication with httpOnly cookies
• QR code contact exchange
• Geocoding and interactive maps
• Parity testing against the original functions
• Security hardening with owner-scoped queries
• Seeded bootstrap intervals
• Paired bootstrap and exact sign test
• LLM vs rule-based baseline evaluation
• BYOK AI with in-browser redaction

• ·A paired study on 980 random boards compared the Manhattan and Euclidean heuristics with a paired bootstrap, Wilcoxon and exact McNemar tests: Manhattan expands fewer nodes but finds a shortest path less often.
• ·Two optional bring-your-own-key features: an LLM Arena where the visitor's model plays the agent, with every move validated against the legal set and results compared game by game with simple baselines, and a move commentator whose claims are checked against the agent's own search. Every call is kept in a local audit log.

Original

Python 3.6 · NumPy · SciPy · Jupyter

2026 revival

Shared stack + Web Workers · zod · Anthropic SDK

Skills

• A* search with Manhattan and Euclidean heuristics
• Minimax with alpha-beta pruning
• Evaluation function design
• Parity testing of a TypeScript port against Python
• Wilson and bootstrap confidence intervals
• Bradley-Terry strength ratings
• Wilcoxon signed-rank and exact McNemar tests
• LLM-as-a-player evaluation
• BYOK AI with a local audit log

• ·Re-checking in 2026 found that the 2023 validation split leaked copied machine texts. Re-fitted on a clean split, the domain-1 detector still separates the classes well, but a text-length baseline comes close, and both are reported with 95% intervals.
• ·A class-imbalance sandbox compares SMOTE with random under-sampling and shows what the notebooks' SMOTE step actually did, and an experiment journal records the meetings, the approaches tried and every leaderboard submission.

Original

Python · Jupyter · scikit-learn · imbalanced-learn · PyTorch · mpi4py · pandas

2026 revival

Shared stack

Skills

• Bag-of-words features on anonymised token ids
• Logistic regression and linear SVMs (SGD)
• Class imbalance: SMOTE and random under-sampling
• CNN experiments (ResNet, ShuffleNet) on count vectors
• Leak-free held-out re-evaluation
• Calibration, AUC and paired tests with intervals
• Exporting model weights for client-side scoring

• ·The browser benchmark repeats each worker count in shuffled rounds and reports medians with exact order-statistic intervals and bootstrap intervals, with coverage checked by simulation.
• ·An optional Ask page answers only from the result tables, must cite rows and show its arithmetic, and every answer is checked automatically and logged locally. A 24-question harness, including questions the tables cannot answer, scores a model on the visitor's own key.

Original

Python 3.7 · mpi4py · Slurm · Spartan HPC · polars · pandas · NumPy

2026 revival

Shared stack + Web Workers · d3-geo · zod · Anthropic SDK

Skills

• MPI parallel processing (scatter and gather)
• Byte-range file chunking
• Regex stream parsing
• Amdahl's and Gustafson's laws
• Benchmark design with repeated randomised rounds
• Order-statistic and bootstrap intervals
• Interval coverage simulation
• Grounded question answering with citation checks
• LLM evaluation with unanswerable questions
• BYOK AI audit logging

• ·Added uncertainty and spatial statistics: t-based intervals per area, Moran's I with LISA cluster maps, and suppression of areas with fewer than 30 tweets. Both headline relationships are consistent with no association.
• ·Optional text-to-SQL on the visitor's own key: the server validates and runs one read-only query, answers must cite rows, and a 16-question harness measures execution accuracy.

Original

Python · mpi4py · NLTK (VADER) · Flask · React 18 · Plotly · CouchDB · Ansible · Docker Swarm · Melbourne Research Cloud

2026 revival

Shared stack + MapLibre GL · SQLite (libSQL) · node-sql-parser · Web Workers · zod

Skills

• Sentiment analysis with VADER
• MPI data processing
• CouchDB MapReduce views
• Spatial autocorrelation (Moran's I, LISA)
• OLS with HC3 robust standard errors
• Bootstrap intervals
• Small-area suppression
• Text-to-SQL with server-side validation
• LLM evaluation (execution accuracy, exact McNemar)
• BYOK AI audit logging

• ·Visitors can shape per-arm Pareto delays, packet loss and reward-dependent lags, then scrub through the confidence bounds, eliminations and posterior draws inside each algorithm.
• ·An optional bring-your-own-key experiment lets a language model choose the arms, scored against Thompson sampling and UCB1 on the same seeds, with every call written to an exportable audit log.

Original

Python · NumPy · SciPy · Matplotlib · Jupyter

2026 revival

Shared stack + Web Workers · WebAssembly · KaTeX · zod · Anthropic SDK

Skills

• Successive-elimination bandits under delayed feedback
• Thompson sampling with Gaussian priors
• Doubly-Adaptive Thompson Sampling
• Offline replay evaluation with delayed rewards
• Bit-exact port of NumPy's PCG64 generator
• Bootstrap intervals over seeded repetitions
• Paired comparisons and effect sizes
• LLM-as-a-policy evaluation
• BYOK AI with a local audit log

• ·Added model checks the coursework did not ask for: simulated residual envelopes, influence, over-dispersion, profile-likelihood intervals, a proportional-odds test and GEE sensitivity, every simulation seeded. Slips in the 2023 write-up are shown beside the recomputed value.
• ·An optional bring-your-own-key reading aid explains each page from its numeric summary only, flags numbers it cannot trace and keeps an exportable audit log. The sensitive Assignment 1 survey appears only as model summaries.

Original

R · R Markdown · nnet · MASS · geepack · ggplot2

2026 revival

Shared stack + KaTeX · zod

Skills

• Logistic and binomial GLMs (dose-response, LD50)
• Poisson log-linear models for contingency tables
• Multinomial and proportional-odds models
• GEE with robust sandwich errors
• Stepwise model selection and analysis of deviance
• Simulated residual envelopes and influence checks
• Porting IRLS and Newton-Raphson fits to TypeScript
• BYOK explanations with automated number checks

• ·The 2026 upgrade adds Wilson intervals on simulated rates and error rates, paired tests wherever two methods share the same data, repeated cross-validation and coverage checks.
• ·A known bug in the 2023 model-selection code can be toggled on and off, and the bootstrap page corrects the submission's claim that the bias-corrected estimator is unbiased. An optional bring-your-own-key AI explainer keeps an audit log.

Original

R · R Markdown · glmnet · e1071 · ISLR · ggplot2

2026 revival

Shared stack + Web Workers · KaTeX · D3 · zod

Skills

• Ridge and lasso regularisation paths
• Cross-validated tuning (cv.glmnet, e1071::tune)
• AR order selection with information criteria
• Monte Carlo simulation studies
• Multiclass SVMs (linear and radial kernels)
• Bootstrap bias correction
• Bit-exact port of R's Mersenne-Twister seeding
• Wilson intervals, McNemar tests and coverage checks

• ·Added a four-chain Bayesian workflow (R-hat, ESS, MCSE, posterior predictive checks, prior sensitivity), which showed the HMC and EP encoding bug passes every convergence check and is caught only by a predictive check.
• ·An optional 'Explain these diagnostics' feature runs on the visitor's key and flags numbers or verdicts that contradict the R-hat and ESS thresholds.

Original

R · R Markdown · knitr · mvtnorm · coda

2026 revival

Shared stack + KaTeX · Web Workers · zod

Skills

• Bayesian logistic regression
• Laplace approximation
• Metropolis-Hastings
• Hamiltonian Monte Carlo
• Expectation propagation
• Convergence diagnostics (rank-normalised R-hat, bulk and tail ESS, MCSE)
• Posterior predictive checks
• Prior sensitivity analysis
• Porting R's RNG and QUADPACK integration
• BYOK diagnostics explainer with automated checks

• ·Fixing the LOOCV and test-set scaling bugs in the rainfall classifiers brings the logistic regression and the classification tree down to one test error out of 41 (the random forest stays at three), and late-January rainfall turns out to be the most informative stretch of the year.

Original

R · R Markdown · MASS · randomForest · pls · rpart · ggplot2

2026 revival

Shared stack + KaTeX · Web Workers

Skills

• Covariance matrices and eigendecomposition
• Principal component analysis
• Partial least squares
• PCA with logistic regression and hand-written LOOCV
• Random forests tuned by out-of-bag error
• Classification trees (rpart)
• Porting R's glm.fit, kernel PLS and rpart to TypeScript

• ·Re-ran all five Hangman guessers on one common split so they are compared pair by pair, with bootstrap intervals on each difference.
• ·An optional LLM Hangman harness plays a model on the same seeded words under the same scoring, counts invalid guesses, refusals and cost separately, and logs each call in the browser.

Original

Python 3.8 · Jupyter · NLTK · scikit-learn · NumPy

2026 revival

Shared stack + Web Workers · Playwright · zod

Skills

• MaxMatch hashtag segmentation
• Unigram and character n-gram language models
• Naive Bayes and logistic regression text classification
• Bootstrap and Wilson intervals
• Exact McNemar test
• Calibration (reliability diagrams, ECE)
• Paired benchmarking on a common split
• LLM evaluation harness
• BYOK AI with a local audit log

• ·It also showed that the committed notebook differs from the code behind the reported results, and that a stray vocabulary term padded the tags of 14% of evidence passages. All three retrieval variants are reproduced: the submitted scoring rule, the notebook's version that counts similarity twice, and its final cell that passes raw text instead of stems.
• ·The weak spots are stated plainly: retrieval found the right passages for only 13 of the 154 development claims, and the retrained classifier never predicts two of the four verdicts, the class-imbalance problem the report warned about.

Original

Python · Google Colab · pandas · NLTK · scikit-learn · PyTorch

2026 revival

Shared stack + SQLite · Recharts · zod

Skills

• TF-IDF keyword extraction and cosine-similarity retrieval
• Text preprocessing (contractions, stopwords, Porter stemming)
• Transformer classifier trained from scratch
• Evidence F-score, accuracy and harmonic-mean evaluation
• Parity-tested TypeScript port down to NLTK's stemmer
• Read-only SQLite passage index