Skip to content
View RizgarOzan's full-sized avatar
💻
we on grind
💻
we on grind
2 followers · 5 following

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Block or report RizgarOzan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
RizgarOzan/README.md
Rızgar Ozan — AI systems and game tools programmer. I build measurement systems: retrieval evals, and game tools.

Website LinkedIn itch.io Email

I build measurement systems: retrieval evaluations that say which part of a RAG pipeline earns its cost, and game tools that score a level or catch a broken font before anyone ships it. Final-year Computer Education & Instructional Technology student at Hacettepe University.

Selected work

Five projects, in the order I'd show them to you. Every number on a card is copied from the project's own committed results.

01 Turkish RAG Eval — 3 chunkers by 4 retrievers on a hand-labelled Turkish gold set. Best nDCG@10 per retriever: BM25 0.41, BM25 with 5-character prefix 0.51, dense 0.50, hybrid RRF 0.61.
02 Match3 Lab — C# match-3 rules core, readable level format, Unity editor tools and a bot simulator. The greedy bot's win rate falls from 100% on level 1 to 50% on level 6; the random bot from 70% to 3%. 1000 games per level. 03 TMP Glyph Audit — Unity package that finds every TextMeshPro text the fonts cannot draw, through TMP's real fallback chain. Editor window plus CI exit codes.
04 Bilim Dedektifi — short educational mystery game, team course project at Hacettepe, free on itch.io. 05 Code-Enigma — React 19 academic progress platform built with İlkhan Arda Akmaca; I built the dashboard, peer comparison markers and the filter and settings controls.

Turkish RAG Eval: 58 hand-labelled queries; differences under 0.05 nDCG are noise, and the README says so. Match3 Lab plays in the browser and ships 50 tests and a decision record for every rule. TMP Glyph Audit: 24 xUnit + 5 EditMode tests, on OpenUPM.

Open source

Pull requests to projects I don't own — the AI and Unity ones listed below, the rest counted. Refreshed every week by a workflow.

19 merged across 7 projects · 17 in review

Pull request Project
🟣 merged fix(markdown): treat an escaped pipe as cell text, not a column separator HKUDS/LightRAG · ★ 39.7k
🟣 merged fix(parser): emit LaTeX-valid delimiters for Word equations HKUDS/LightRAG · ★ 39.7k
🟣 merged fix: pass the selected benchmarks in create-model-results embeddings-benchmark/mteb · ★ 3.4k
🟣 merged fix: apply the n_parameters lower bound when there is no upper bound embeddings-benchmark/mteb · ★ 3.4k
🟣 merged Fix compiled models with losses that require media counts in the trainer huggingface/sentence-transformers · ★ 19.1k
🟣 merged fix: reopen existing NumpyCache without a -1 memmap shape embeddings-benchmark/mteb · ★ 3.4k
🟣 merged Create com.rizgarozan.tmp-glyph-audit.yml openupm/openupm · ★ 1.9k
🟣 merged fix(auth): strip whitespace around AUTH_ACCOUNTS entries HKUDS/LightRAG · ★ 39.7k
🟣 merged fix(docx): record Roman and ideographTraditional counts past their domain HKUDS/LightRAG · ★ 39.7k
🟢 in review fix(parser): make filename-hint validation and classification agree HKUDS/LightRAG · ★ 39.7k

…and 10 more merged pull requests to mdn, radzenhq.

Toolbox

Unity C# / .NET Python React GitHub Actions Git

Header and cards are SVGs rendered by scripts/ in the site's palette; the cards' text is outlined so they look the same on every machine.

Pinned Loading

  1. tmp-glyph-audit tmp-glyph-audit Public

    Find every TextMeshPro text your fonts cannot draw - scenes, prefabs and runtime text, through TMP's real fallback chain. Editor window + CI exit codes.

    ShaderLab 1

  2. turkce-humanizer turkce-humanizer Public

    Türkçe yapay zekâ metnini, ne dediğini değiştirmeden yazarın sesine çeviren ajan skill'i (Claude Code, Codex, Cursor). Humanizer'ın Türkçe karşılığı.

    Python 1

  3. match3-lab match3-lab Public

    Match-3 workbench: pure C# rules core, readable level format, bot simulator that measures difficulty, Unity editor tools + WebGL demo

    C#

  4. turkish-rag-eval turkish-rag-eval Public

    Which parts of a RAG pipeline earn their cost in Turkish? 12 retrieval configs measured on a hand-labelled Turkish gold set

    Python 1 1