Skip to content
View amkkoussay's full-sized avatar

Block or report amkkoussay

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. 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
amkkoussay/README.md
Koussay Mehdouani — programmer and poet Applied AI for real problems — honesty engineering

Portfolio LinkedIn Zenodo paper Gmail

Open to training Bac 2028 Focus: applied AI and data Based in Tunisia


Waving hand Hi, I'm Koussay

Rotating Earth

I'm a Tunisian high-school student (Sciences de l'Informatique, bac 2028) building applied-AI and health tools that know their limits. I work on applied AI, health informatics, agriculture, and climate risk — small systems with clear limits and readable behavior.

🔬 Honesty engineering — the boy who teaches machines to say: I don't know. My research and my projects ask the same question: what does a system owe the people who trust it, when its data cannot be trusted?



🛠️ My toolbox

Python, JavaScript, Flask, React, Vite, SQLite, Git, GitHub, VS Code, HTML, CSS

Python · JavaScript · Flask · React · Vite · SQLite · Git — small stack, shipped often.


🚀 Featured projects

⭐ Top 3 — the flagship builds

MIRSAD icon
MIRSAD · مِرصاد
One shared memory for every AI assistant — facts, decisions with their reasons, and conflicts, served over MCP.

MIRSAD demo — live MCP tool calls on fictional demo data
Rain2Risk icon
Rain2Risk
Transparent rainfall-risk screening for any point on Earth — a score that prints its own data limits. Live demo.

Rain2Risk demo — a live screening recorded on the production site
CropGuide icon
CropGuide AI
A plant-health field guide in Arabic, French and English — designed to say “I don't know” when the evidence is not enough.

CropGuide demo — the app running, including the Arabic interface
  • 🔭 MIRSAD — مِرصاد — one shared, living memory for every AI agent: bi-temporal facts, decisions stored with their reasons and evidence, and a cross-agent conflict queue only the owner resolves — served over MCP with a live graph view. Local-first; zero LLM in the write path.
  • 🌧️ Rain2Risk — transparent relative flood-risk screening with documented data limits and tests. Live: rain2risk.vercel.app
  • 🌾 CropGuide AI — agricultural decision support with explicit uncertainty, data audits, and tests.

Honest limits, printed on the label: both are exploratory software. CropGuide AI is not a substitute for agricultural expertise; Rain2Risk is a screening score, not an official warning or forecast.

📂 More builds — click to open
Project What it is
Bakery ERP Pro Bakery management demo — stock, sales, production (a demo project, not production ERP)
AI Health Assistant Health-assistant prototype built on careful, limited claims
Clinique Clinic management system
Finance Platform Personal-finance tracking platform

📄 Research spotlight

AI-Based Clinical Decision-Support System Robustness Under Unreliable Patient Data — a narrative review and thematic synthesis, published as a preprint on Zenodo in 2026. Fifteen studies, six robustness themes, one conclusion: a clinical AI that cannot say "I don't know" is not safe yet.

DOI 10.5281/zenodo.22883408


🤝 Open Source Contributions

Merged work in established projects — every entry links to the merged pull request.

  • pandas — fixed a silent data bug: pd.to_numeric(..., errors="coerce") on Arrow-backed strings stored a plain float NaN instead of pd.NA, so the missing values escaped isna() — and next to an existing missing entry, a neighbouring value could be duplicated in their place. Root-caused inside to_numeric, fixed with regression tests, and merged into main for pandas 3.2.0, where I am credited in the release notes. → PR #71991 · closes #67949
  • Microsoft Azure documentation — two merged fixes in Microsoft's official Azure docs: a duplicated word in the Azure Monitor metrics article (PR #128901) and in the Blob Storage CLI quickstart (PR #128902).

Currently under review: curl · scikit-learn · xarray · NVIDIA · lighteval


🏆 Competitions

I compete on Zindi as koussayy — ranked #7,237 among 100,000+ AI builders (best rank #7,208) · 49 points · 56 submissions across 5 challenges (standings as shown on Zindi, 10 Nov 2026).

🟢 Active

  • Financial Stress Prediction Challenge (Beginner Track, October Edition) — predicting financial stress from survey data, scored on log loss + ROC-AUC. Rank #56 of 196, best public score 0.7360881, 25 submissions. Closes 29 Nov 2026.

✅ Completed

  • Bias Bounty: Mapping Equity Challenge (Humane Intelligence × Radiant Earth) — estimating Overture Maps coverage gaps across 9,794 US census tracts with a fully documented, reproducible, no-training methodology: a rule-based index, a wrong denominator found and fixed, then shrinkage toward regional means. Final standing #132 of 505, private score 0.085128854 (public 0.081706769), 31 submissions. Closed 1 Nov 2026; under code review, with rankings provisional until 22 Nov 2026.

📅 Joined — upcoming

Zindi badge — First Steps (first submission) Zindi badge — Go-Getter (profile completed) Zindi badge — Zindian (verified profile)

First Steps · Go-Getter · Zindian

Zindi profile — koussayy Financial Stress — rank #56 Best public score 0.7360881


📊 The numbers, live

GitHub stats Top languages

Contribution streak

Profile details and contribution graph

Contribution chart


🎓 Certificates I'm proud of

🏅 Verified highlights — click to open
  • CS50AI — Introduction to AI with Python · Harvard / CS50 — verify
  • Elements of AI · University of Helsinki / MinnaLearn (2 ECTS) — verify

Certificates

  • AI in Healthcare Specialization · Stanford Online (Coursera)
  • AI in Government · Oxford Saïd

Developer at work

A quote to build by

Thanks for stopping by — the next build is already on the desk. · © 2026 Koussay Mehdouani

Pinned Loading

  1. cropguide-ai cropguide-ai Public

    CropGuide AI — take a photo of a sick plant and see possible disease names. 40 crops, 439 diseases. In Arabic, French and English. Not a final diagnosis.

    JavaScript 1

  2. rain2risk rain2risk Public

    Python 1

  3. Mersad Mersad Public

    MIRSAD (مِرصاد) — a shared, living memory hub for AI agents: bi-temporal facts, first-class decisions, owner-resolved conflicts, and a live graph view. Local-first

    Python 13

  4. pandas-dev/pandas pandas-dev/pandas Public

    Flexible and powerful data analysis / manipulation library for Python, providing labeled data structures similar to R data.frame objects, statistical functions, and much more

    Python 50.6k 20.8k

  5. scikit-learn/scikit-learn scikit-learn/scikit-learn Public

    scikit-learn: machine learning in Python

    Python 68.2k 27.7k

  6. NVIDIA/structured-data-models-client NVIDIA/structured-data-models-client Public

    One Python client for NVIDIA's structured-data foundation models: ask a predictive question of one table or of a whole relational schema, and get an answer without training anything.

    Python 13 6