01 · Roasts
Portfolio, not pull requests
Eight named projects and 524 yearly commits, but 0 PRs, 0 issues, 0 stars, and 0 forks: the shipping is real; the outside feedback loop is not.
CI is selective
tickforge and Polymarket_bot automate checks, while Macro-Bias, Identity, and the flagship portfolio still leave reliability to optimism.
The zero-star paradox
Macro Bias has pricing routes, TickForge has a live Explorer, and Toolbox has a domain—yet the dashboard still reads 0 stars across the board.
Horizontal builder detected
166 recent commits sampled across repos and a 232 MB Polymarket_bot say this is far beyond a one-repo tutorial arc.
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% weight62C
- Consistency20% weight65C
- Quality20% weight71B
- Depth15% weight58D
- Breadth10% weight80A
- Community10% weight25F
03 · Stats
365-day commit heatmap
68 active days
Language distribution
- Python70%
- TypeScript20%
- Rust7%
- C++1%
- Shell0%
- CSS0%
- Other2%
04 · Numbers
Owned repos
non-fork
9
Commits
last 12 months
524
Followers
2
Joined GitHub
Aug 2024
05 · Top repos
FiNdAlMkSkInDaL /
tickforge
A documented C++23 limit-order-book project with matching, self-trade prevention, Avellaneda–Stoikov simulation, browser demo, tests, benchmarks, and cross-platform CI; adoption is currently unproven at 0 stars.
FiNdAlMkSkInDaL /
Polymarket_bot
A substantial, documented Polymarket shadow-research stack with live L2 Parquet capture, backtesting, multi-signal orchestration, risk guards, hourly lanes, and Telegram telemetry, but no visible adoption or licensing.
FiNdAlMkSkInDaL /
Macro-Bias
Typed Next.js/Supabase macro-regime product with stocks and crypto KNN research, walk-forward evaluation, paper ledgers, publishing integrations, and substantial schema-backed application structure; adoption remains unproven at 0 stars.
FiNdAlMkSkInDaL /
latent-control-lab
A well-documented research demo that routes frozen distilgpt2 activations through a tested linear probe into a sandboxed 5×5 VectorBot, but has no visible adoption or external product evidence.
FiNdAlMkSkInDaL /
FiNdAlMkSkInDaL.github.io
A deployed, highly designed GitHub Pages portfolio that bundles six interactive project demos, shareable routes, CV downloads, and a polished pixel-garden interface, but has no visible external adoption or automated tests.
FiNdAlMkSkInDaL /
Identity
A substantial, well-documented Rust Windows prototype implementing local capture, SQLite transit processing, protected storage, memory graphs, vector backends, and scoped context generation, but with no visible adoption, CI, or license.
FiNdAlMkSkInDaL /
asof
A focused, well-documented Python reconciliation tool with cassette replay, SQLite state, a rule-driven policy, and extensive tests, but no visible adoption, CI, or typed-language status.
FiNdAlMkSkInDaL /
FiNdAlMkSkInDaL
A maintained personal portfolio README linking five named engineering projects and a GitHub Pages site, but this repository contains no implementation, tests, CI, license, or typed source of its own.
06 · Timeline
- Aug 7, 2024Joined GitHub
- Feb 27, 2026Created Polymarket_bot — Live Polymarket L2 lake + hourly shadow strategies on a small VPS. Research/learning, not live auto-trading.
- Apr 5, 2026Created Macro-Bias — Daily macro/crypto regime research app: walk-forward scores, briefings, paper paths. Learning project.
- Jun 1, 2026Created Identity — Local-first Rust context daemon: capture, memory graph, narrow clipboard context. Windows prototype.
- Jun 8, 2026Created latent-control-lab — VectorBot: route frozen LM activations to grid actions with no text generation. Toy research demo.
- Jun 20, 2026Created FiNdAlMkSkInDaL — README
- Jul 13, 2026Created tickforge — C++ limit order book + matching + simple Avellaneda-Stoikov MM. Learning project, not a trading product.
- Jul 13, 2026Created FiNdAlMkSkInDaL.github.io — Personal portfolio site
- Aug 13, 2026Created asof — Reconcile a live Polymarket order book with a Gamma warehouse snapshot
- Aug 18, 2026Most recent push to FiNdAlMkSkInDaL.github.io
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.