flyrank-ml-internship-starter

Finds label leakage and designs honest validation — leakage taxonomy, grouped and time-aware splits, base rates, and the attack-your-own-model checklist. Use before trusting any metric, when a score looks too good, or when features and labels share time windows or origins.

The FlyRank internship datasets — the 30k-row starter CSV and its gotchas, the ~79M-row warehouse release tables and grains, panel warnings, access, and iteration rules. Load for EVERY task that touches the data. (Project-specific: delete this folder when reusing the skill library elsewhere.)

Works with datasets far too big to download or load in pandas — SQL over remote Parquet with DuckDB, aggregate-then-model, iterate on samples. Use when a dataset has millions of rows, lives on a remote host (hf:// or s3), or a notebook runs out of memory or hits rate limits.

Runs an honest EDA and signal audit — distributions first, heavy-tail handling, verdict-style mini-tests (CONFIRMED/OPPOSITE/MIXED/FALSE), sample-size floors. Use before modeling, when testing whether a popular belief shows up in data, or when a correlation looks too good.

Builds the transparent rule-based baseline every model must beat — a hand-written score with reason codes, ranked output, and precision@K evaluation. Use before training any model, or when someone reports model results with nothing to compare against.

Deploys a static page (research paper, portfolio piece) for free from a GitHub repo using GitHub Pages — setup, file layout, verification, and recording the final URL. Use when publishing a finished write-up as a public web page.

Teaches the working method for using an AI coding assistant on analysis tasks — one task per fresh conversation, search before assuming, interview before big plans, validate everything, commit per finished task. Use at the start of any assignment, whenever a conversation gets long and confused, or when the assistant starts guessing.

FlyRank-specific background — what the company does, the search problem, and the hand-written flags story that the ML internship work builds on. Load for context when an assignment references FlyRank's product, flags, or the "beat the rule" framing. (Project-specific: delete this folder when reusing the skill library elsewhere.)

Frames a data/ML problem before any modeling — the decision, the action, the cost of a wrong call, task type, target, and success metric. Use when picking a project direction, writing a research question, or mapping a problem onto an ML task type.

Trains a first model the honest way — method chosen to fit the question, compared against the baseline on the same split and metric, errors read before scores are believed. Use when moving from a rule baseline to a learned model, or when reviewing a model that reports only a single score.

Writes a data contract — what a row means, which fields are features vs labels vs context vs excluded, over which time windows — and verifies every claim with a query. Use before any feature building or modeling, or when results look wrong and the data definition is suspect.

Writes findings in language the evidence can carry — the claim ladder (observed → directional → decision-support, never causal without a design), effect sizes over drama, banned phrasings. Use when writing any conclusion, report section, playbook, or public page from analysis results.

Structures and writes a public research page — canonical sections (abstract through limitations), storytelling that carries findings, chart hygiene, referencing for credibility, repurposing for different audiences. Use when turning finished analysis into a paper, portfolio page, or public write-up.