01 · Roasts
CI/CD? Never heard of her.
7 repos scored. HAS_CI=yes count: 0. HAS_TESTS=yes count: 0. You're out here implementing Park et al.'s phantom sensation algorithm from scratch but can't phantom a pytest file into existence.
The Weekend Warrior
Hack2026 — 635 MB of robotics code, 10 commits, 2 days. Either you're a genius or you git-added a conda environment. The absence of a .gitignore suggests the latter.
Profile repo doing the heavy lifting
DariusGiannoli repo: 11 KB, 11 commits over a year, sole content is your own name. That's a commit rate of ~1/month to maintain a README that says 'Darius Dayu Giannoli'. Dedicated.
Soloist (98%)
soloPct = 98. You've opened 20 PRs this year on other people's code, which is great — but in your own repos you are, statistically, talking to yourself. At least you're a good listener.
Burst Coder in a Streak Economy
Your heatmap is a seismograph, not a heartbeat — weeks of silence punctuated by frantic 4-commit days. 451 commits/year is solid, but GitHub rewards the tortoise, not the caffeinated sprinter.
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% weight56D
- Consistency20% weight60C
- Quality20% weight52D
- Depth15% weight55D
- Breadth10% weight65C
- Community10% weight40D
03 · Stats
365-day commit heatmap
137 active days
Language distribution
- Python50%
- Jupyter Notebook29%
- TypeScript6%
- C#4%
- MDX4%
- Java4%
- Other3%
04 · Numbers
Owned repos
non-fork
11
Commits
last 12 months
451
Followers
5
Joined GitHub
Dec 2023
05 · Top repos
DariusGiannoli /
LIS
Tactile hardware/motion pattern research platform with multi-phase experimental design (A/B phases), typed Python core implementing Park et al. phantom algorithms, serial/BLE device APIs, and GUI interfaces. Non-trivial project architecture spanning multiple protocol implementations and study interaction patterns.
DariusGiannoli /
SwarmControl
Research implementation for body-based drone swarm control in VR using IMUs and hand tracking, built in C#/Python with calibration tools. Typed Python modules with structured src/ layout, meaningful docs in README, but no tests/CI and limited external adoption signals.
DariusGiannoli /
IntroToMachineLearning
Educational ML algorithms implementation: KNN, logistic regression, linear regression, kernel methods, SVM with clean typed Python, structured src/ layout, and evaluation utilities. Active course project with 9 commits in ~11 days but no README, tests, or CI.
DariusGiannoli /
RecognitionBenchmark
Jupyter notebook + Streamlit web app benchmarking RCE feature extraction vs modern NN models (ResNet, MobileNet, YOLOv8) on bird detection. Typed Python with structured multi-file layout, meaningful docs in README + app.py, but no tests, CI, or license. ~50 commits over 7 weeks with 15MB codebase shows sustained develo
DariusGiannoli /
Hack2026
Early-stage robotics teleoperation project with 635 MB codebase, minimal documentation (README only), no tests/CI/typing, 10 commits in 2 days. Experimental hardware-software integration work.
DariusGiannoli /
Student_Survey
Academic survey analysis project for EPFL course with Python data analysis script, CSV dataset, minimal docs, no tests/CI/license, and 6 commits over 3 months.
DariusGiannoli /
DariusGiannoli
Minimal GitHub profile config repository with a trivial README and no source files. Created in Feb 2025, shows 11 commits over the last month but represents a placeholder profile setup rather than a substantive project.
06 · Timeline
- Dec 4, 2023Joined GitHub
- Feb 27, 2025Created DariusGiannoli — Config files for my GitHub profile.
- Jul 21, 2025Created LIS
- Nov 21, 2025Created SwarmControl — Upper Body Drone Swarm Control - EPFL Laboratory of Intelligent Systems
- Dec 1, 2025Created Student_Survey
- Feb 14, 2026Created RecognitionBenchmark — Comprehensive benchmarking of RCE NN with feature extraction compared to industry standard NN
- Mar 20, 2026Created Hack2026
- Apr 13, 2026Created IntroToMachineLearning
- Apr 29, 2026Most recent push to SwarmControl
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.