01 · Roasts
Burst Builder Extraordinaire
12 repos, most created and last-pushed within the same 24–48 hours. acetone-iodination: created Feb 11, last push Feb 12. electrochem-thermo-lab: created AND last pushed Feb 27. You ship fast and then disappear like a lab technician at 5pm.
Documentation Enthusiast, Test Avoider
design.md, ARCHITECTURE.md, STATUS.md — you've got the whole documentation family except the one that actually catches bugs. Not a single repo has CI, and only 2 out of 12 have any tests at all. The docs are immaculate; the coverage is 0%.
The 25-Impact Ceiling
Every single repo scored exactly 25 on Impact. That's not consistency — that's a floor. Zero external stars across 11 of 12 repos, 1 lonely fork on continuum-discovery. You're building in a completely empty room.
README Selective Service
pharmlink has zero README. aero-kinetic has zero README. Both have fancy architecture and typed frontends. You spent time on TanStack Router and Radix UI components but couldn't open a markdown file and write 'what this does.'
The Prolific Scaffolder
continuum-discovery: 'satellite pathogen detection, protein engineering, biodefense swarm' — 30 commits, 10 days, 0 lines of actual implementation visible. sikaflow: autonomous financial coordination framework, wallets are hardcoded test strings, execution defaults to simulation. The vision is S-tier; the code is a README in a trenchcoat.
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% weight68C
- Consistency20% weight65C
- Quality20% weight57D
- Depth15% weight58D
- Breadth10% weight65C
- Community10% weight55D
03 · Stats
365-day commit heatmap
54 active days
Language distribution
- Python52%
- Jupyter Notebook28%
- Shell7%
- TypeScript6%
- HTML3%
- JavaScript2%
- Other2%
04 · Numbers
Owned repos
non-fork
25
Commits
last 12 months
459
Followers
18
Joined GitHub
Aug 2022
05 · Top repos
pkaysantana /
SikaSentinel
TypeScript policy + audit engine for Hedera financial workflows; well-typed, structured modules, deterministic NLP parser, and comprehensive integration tests—but brand new (3 days old), no README, and zero adoption signals.
pkaysantana /
arctreasury
TypeScript/Solidity treasury engine with deterministic USDC payout controls, RBAC, and O(1) yield distribution. Typed, tested, well-documented—but brand new (4 days old), 0 stars, focused on Arc Layer-1 hackathon submission.
pkaysantana /
GEM-Adaptyv-RBX-1-2026
Competition submission for RBX-1 protein binder design using RFdiffusion + ProteinMPNN pipeline, with 25 validated sequences ranked by computational metrics (pair_iptm_AB, pLDDT). Well-documented four-stage pipeline with multiple CSV metrics across intermediates and final submission, but no tests or CI. Personal resear
pkaysantana /
pharmlink
PharmLink is a TypeScript React demo app for pharmacy surplus matching, built in ~24 hours. Typed, structured with TanStack Router, clean UI with Tailwind & Radix. Zero tests, no CI, no docs, but solid MVP-grade architecture targeting a real healthcare use case.
pkaysantana /
perovskite-band-structure-simulator
Specialized computational chemistry project simulating perovskite band structures; well-documented via docs/, design.md, ARCHITECTURE.md, and STATUS.md despite no README; typed Python absent but structured documentation and scientific scope evident.
pkaysantana /
acetone-iodination-analysis
Personal chemistry lab automation project: Python pipeline for kinetic analysis of iodination reaction with rate calculation, RANSAC robustness, Hofmeister salt correction, and Streamlit dashboard. Typed YAML config, structured src/ layout, documented in README, but no tests/CI and created 2026-02-11 with only 8 commit
pkaysantana /
aero-kinetic
Educational digital twin for gas viscosity experiments combining Python Poiseuille flow kernel with React "Mission Control" dashboard. Typed frontend with physics engine, but minimal commit history and no tests/CI.
pkaysantana /
keto-metabolic-simulator
Educational biochemistry simulator visualizing pH-dependent ionization of keto-acids using Canvas and Plotly.js. Vanilla JavaScript project with comprehensive documentation (README, design.md, ARCHITECTURE.md, STATUS.md) but lacks tests, CI, and licenses. Created Feb 15–16, 2026 with minimal commit history (9 commits i
pkaysantana /
sikaflow
Ultra-early-stage autonomous financial coordination framework with intent-driven orchestration and policy gates. Well-documented concept but nascent codebase (37 KB, ~5–8 days old) with untyped Python, no tests/CI, incomplete service implementations, and extremely minimal adoption signals.
pkaysantana /
enthalpy-entropy-vaporisation-lab
Educational lab automation for Clausius-Clapeyron analysis with structured code, clear documentation, and mock datasets. Narrow domain scope with no production use or external adoption.
pkaysantana /
electrochem-thermo-lab
Educational Python toolkit for electrochemical cell analysis with thermodynamic calculations. Brand new (hours old), minimal commit activity, no tests or CI, untyped Python code, but has README with clear methodology and MIT license.
pkaysantana /
continuum-discovery
Early-stage Python biodefense swarm project with ambitious vision (satellite pathogen detection, protein engineering), comprehensive documentation (design.md, ARCHITECTURE.md, STATUS.md), but no actual implementation—30 commits in 10 days with sampled files unavailable suggests mostly scaffolding and design.
06 · Timeline
- Aug 23, 2022Joined GitHub
- Dec 4, 2025Created perovskite-band-structure-simulator — Computational chemistry project that connects ion–ligand complexes to extended perovskite ABO3 materials. Simulates metal–ligand coordination, d–p orbital overlap, and simplified b
- Feb 11, 2026Created aero-kinetic — Digital twin for Y1 Physical Chemistry Experiment 1: Determining gas viscosities and molecular diameters. Features a Python physics kernel for Hagen-Poiseuille flow analysis and a
- Feb 11, 2026Created acetone-iodination-analysis — Automated kinetic engine for determining reaction orders, rate constants, and Arrhenius parameters for the acid-catalysed iodination of acetone.
- Feb 15, 2026Created keto-metabolic-simulator
- Feb 25, 2026Created enthalpy-entropy-vaporisation-lab — Automated analysis of vapour pressure data for cyclohexane and methanol using the Clausius-Clapeyron equation to determine $\Delta_{vap}H$, $T_b$, and $\Delta_{vap}S$.
- Feb 27, 2026Created electrochem-thermo-lab — Dual-method analysis of galvanic cells: Temperature-dependence of E_cell and calorimetric validation of ΔH.
- Feb 28, 2026Created arctreasury
- Mar 1, 2026Created continuum-discovery
- Mar 3, 2026Created GEM-Adaptyv-RBX-1-2026
- Mar 21, 2026Created sikaflow
- Apr 9, 2026Created SikaSentinel
- Apr 19, 2026Created pharmlink — PharmLink is a PMR-integrated workflow layer for community pharmacies that surfaces nearby surplus stock at the point of reorder, enabling auditable local transfers before defaulti
- Apr 19, 2026Most recent push to pharmlink
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.