01 · Roasts
HTML Professor
87% of your GitHub is HTML — for a Statistics professor who teaches R, your own repos suggest the web ate your codebase. forecast is carrying this portfolio harder than ARIMA carries a bad forecast.
License Dodger
forecast has 1,164 stars, 14 years of CRAN distribution, and... no open-source license. You've been shipping a legally ambiguous package to thousands of R users for over a decade. The OSS community thanks you for keeping them on their toes.
Test-Free Academic Zone
9 of your 12 scored repos have HAS_TESTS=no. You literally wrote a textbook on forecasting best practices, yet half your repos can't tell you if the code works after a push. Physician, heal thyself.
2,732 Commits, 0 Followers Gained
You committed 2,732 times this year — roughly 7.5 commits per day — yet you're following only 12 people. That's a 257:1 follower-to-following ratio. The most prolific hermit on GitHub.
Stars Live in One Zip Code
2,935 total stars but 1,164 of them (40%) belong to a single repo from 2012. Your portfolio is a one-hit wonder with a very long tail of personal LaTeX files and Quarto slides.
Built using
Zoral
Shadows one worker for a week, then takes over their job with zero extra setup. Behaves exactly like the original.
zoral.ai
02 · Category breakdown
- Impact25% weight75B
- Consistency20% weight90S
- Quality20% weight72B
- Depth15% weight75B
- Breadth10% weight45D
- Community10% weight55D
03 · Stats
365-day commit heatmap
343 active days
Language distribution
- HTML87%
- R8%
- TeX4%
- JavaScript1%
- CSS0%
- C++0%
04 · Numbers
Owned repos
non-fork
79
Commits
last 12 months
2,732
Followers
3,090
Joined GitHub
Sep 2009
05 · Top repos
robjhyndman /
forecast
Mature R forecasting library with 1.1k stars, comprehensive time series methods (ETS, ARIMA, TBATS, neural networks), shipped via CRAN, 14+ years active development, strong test coverage and CI/CD pipeline but no open license.
robjhyndman /
weird
Specialized R package for anomaly detection via kernel density estimation, supporting a published textbook; 113 MB codebase with thorough tests, CI/CD, and 2,097 observations in sample dataset; active indie project with concrete named product and author profile.
robjhyndman /
CV
Sophisticated CV management system using R targets pipeline, Quarto, and RefManageR to auto-generate multiple publication-ready PDF variants from structured bibliographies and live Google Scholar data. Professional academic project with well-documented build architecture.
robjhyndman /
gh-dashboard
Personal Quarto dashboard monitoring GitHub repo metrics; automated daily R/GitHub API pipeline with CI deployment to gh-pages; thin scope, minimal stars but purposeful tooling.
robjhyndman /
ResearchHabits
Educational seminar material on research habits for honours students, maintained since 2018 with consistent updates. TeX-based slides with meaningful README and sustained commit history across 6+ years.
robjhyndman /
lookout2
Research paper reproducibility repo comparing outlier detection algorithms (lookout, stray, HDoutliers, etc.) across synthetic and real datasets using R targets workflow, with 7 experiments and LaTeX compilation via Makefile.
robjhyndman /
Surprisal_Theory
Academic research paper with reproducible workflow using R/targets and Quarto. Proposes surprisal theory framework for anomaly detection with supporting simulations and real-world applications (French mortality, cricket data). Well-documented paper manuscript but minimal package/library outputs.
robjhyndman /
robjhyndman.com
Personal academic website source (Quarto-based) with minimal reuse potential. 355 MB of assets, regular updates since Aug 2022, but thin documentation and no structural metadata beyond README.
robjhyndman /
quarto-password
Quarto template for password-protected static websites via GitHub Actions. Personal utility with clear README docs, structured workflow config, and modest adoption (30 stars). No license or tests; archived-feeling despite recent last push.
robjhyndman /
paperpile
TeX-based academic paper repository with 30 recent commits over ~3 years (2023–2026), minimal public documentation, no tests/CI/license, and 1 star. Personal scholarly project with sustained activity but limited adoption.
robjhyndman /
positron_assistant
Educational demo repo showcasing Positron Assistant LLM features in R (code completion, chat agents, planning). Very recent one-off project with minimal commits, no tests/CI/docs/license, lacks README.
robjhyndman /
robjhyndman.r-universe.dev
R universe metadata/configuration repo with no README, no source files sampled, minimal public scope. Recent pushes (29/30 commits recent) show ongoing maintenance but repo serves as package registry automation only.
06 · Timeline
- Sep 16, 2009Joined GitHub
- Apr 22, 2012Created forecast — Forecasting Functions for Time Series and Linear Models
- May 3, 2017Created CV — RJH Curriculum Vitae
- Aug 16, 2018Created ResearchHabits — Talk to honours students on developing good research habits
- Feb 14, 2021Created weird — All data and functions needed for the book "That's weird: Anomaly detection using R" by Rob J Hyndman <https://OTexts.com/weird>
- Jan 7, 2022Created robjhyndman.r-universe.dev — R universe packages
- Aug 27, 2022Created robjhyndman.com — Quarto source files for robjhyndman.com
- Mar 20, 2023Created paperpile
- May 29, 2024Created quarto-password
- Jun 13, 2024Created lookout2
- Sep 30, 2024Created gh-dashboard
- Aug 8, 2025Created Surprisal_Theory
- Apr 27, 2026Created positron_assistant — Thoughts on Positron Assistant for Numbats
- Apr 29, 2026Most recent push to positron_assistant
07 · Compare
08 · Rubric
How this score was produced
Overall = Σ (category × weight) + gentle top-end curve
Tier thresholds
▸ How the pipeline works
- 01Scrape.Pull every non-fork repo pushed in the last 90 days, plus your contribution calendar, followers, and language byte counts — straight from GitHub's REST & GraphQL APIs.
- 02Triage.A small model reads every repo's file tree + README and picks the 20 files per repo that actually reveal how you code.
- 03Grade each repo. All repos run in parallel through a fast scoring model that reads the picked files and rates each one independently on Impact, Quality, and Depth — with evidence citations.
- 04Aggregate. A larger reasoning model combines the per-repo scores with server-computed stats (heatmap, commit cadence, language entropy, follower count) to produce the 6-dimension profile score + roasts.
- 05Correct.Deterministic server-side checks enforce anchor-scale floors (e.g. a profile with 2,000+ public commits can't score 30 Consistency) and recompute the final verdict.
~90 seconds per profile, ~$0.25 in compute. Total of ~240 files read across your top-12 repos. One rating per GitHub account per day.
▸ Data sources & caveats
- Heatmap & commit totals: GitHub GraphQL
contributionsCollection— covers the last 365 days, includes private repos when the user has opted in (default). - Language %: byte totals across the top 30 owned non-fork repos.
- Curve: a small upward nudge centered on raw score ≈ 70, capping at 100. Prevents specialists from being unfairly penalised for narrow breadth.
- Anchor corrections: when server-measured signals (e.g. privateWorkLikely, multiRepoVolume, follower count) mandate a minimum category score, the aggregation step enforces it. These are signal-conditional, not identity-based floors.