Skip to content
View tarekmasryo's full-sized avatar
203 followers · 86 following

Achievements

Achievement: Pull SharkAchievement: YOLO

Achievements

Achievement: Pull SharkAchievement: YOLO

Block or report tarekmasryo

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
tarekmasryo/README.md

Tarek Masryo Banner

Typing SVG

AI/ML Engineer focused on Decision Ops and Observability — building reproducible ML systems and reliability-first GenAI/RAG pipelines.
From raw data → decision-ready insights → deployable systems.

Kaggle Datasets Grandmaster Kaggle Notebooks Expert

GitHub Website Repos LinkedIn

Kaggle Hugging Face Streamlit

🧭 What I do

Area What you can expect
Production ML pipelines Clean data → features → training → evaluation → inference-ready artifacts (repeatable + testable)
Decision Ops dashboards KPI-first UX, drill-down analytics, operating thresholds, cost/capacity trade-offs
GenAI/RAG reliability Schema-first outputs, validation + retries, retrieval evaluation, telemetry for debugging & drift
MLOps & quality CI-friendly delivery, artifact/versioning discipline, monitoring mindset

🌟 Featured portfolio

🧩 Dashboards & Apps

Project Focus Link
Fraud Detection Dashboard Streamlit app integrated with ML artifacts + decision-oriented UX Repo
Streamlit profile Deployed dashboards gallery Profile
Hugging Face profile Spaces + Datasets Profile

🧠 LLM System Ops (Decision-Grade Observability)

Project Focus Link
LLM System Ops — Production Telemetry Telemetry → policies: budget burn, hotspots, routing backtests, drift highlights Repo

📦 Data products (Kaggle)

Dataset What it’s for Link
YouTube Shorts & TikTok Trends 2025 Short-form trends analytics and virality exploration Dataset
RAG QA Evaluation Logs & Corpus Evaluating RAG reliability + QA log analysis Dataset
Cancer Risk Factors Clean features for health EDA and risk modeling Dataset
Football Matches 2024/2025 (Top Leagues + UCL) Standardized match-level data for analytics/modeling Dataset
Digital Lifestyle & Mental Wellness Behavioral signals for wellbeing analytics and prediction Dataset

🧰 Systems & Pipelines

Project Focus Link
Credit Card Fraud Detection — A Pipeline Journey End-to-end pipeline thinking + evaluation mindset Repo
Text Sentiment Analysis NLP workflow, modeling, evaluation structure Repo
Pima Diabetes Pipeline Production-minded pipeline layout (train/evaluate/infer) Repo

🛠️ Tech stack

Category Tools
Languages & Core Python SQL Bash Git Linux
Data & Analytics NumPy Pandas Jupyter Polars DuckDB
ML / DL PyTorch scikit-learn XGBoost LightGBM TensorFlow
Apps & Visualization Streamlit Plotly Matplotlib Seaborn
APIs & Deployment FastAPI Pydantic Docker
MLOps & Quality MLflow GitHub Actions pytest Ruff
GenAI / RAG Hugging Face Transformers Gradio

🤝 Collaboration

  • 📊 Decision Ops: threshold policies, cost/capacity trade-offs, KPI-to-action dashboards
  • 🛠️ Pipeline & MLOps review: reproducibility, artifact/versioning, CI structure, inference packaging
  • 🧠 RAG/LLM reliability: schema-first outputs, validation + retries, retrieval evaluation, telemetry for debugging & drift

Best contact: LinkedIn

If you find the work useful, a ⭐ helps more people discover it.

Footer Banner

Pinned Loading

  1. pima-diabetes-pipeline pima-diabetes-pipeline Public

    End-to-end diabetes risk prediction pipeline (Pima): EDA → feature engineering → calibration + cost-aware threshold → deployable artifacts.

    Jupyter Notebook 9

  2. tarekmasryo.github.io tarekmasryo.github.io Public

    Tarek Masryo — AI/ML Engineer Portfolio

    JavaScript 2

  3. text-sentiment-analysis text-sentiment-analysis Public

    IMDB reviews sentiment analysis: EDA → TF-IDF baselines (NB/LogReg/Linear SVM + calibration) → F1 threshold tuning → explainability → BiLSTM baseline.

    Jupyter Notebook 6

  4. fraud-detection-dashboard fraud-detection-dashboard Public

    Production-minded Streamlit + Plotly fraud detection dashboard with decision policies (Strict/Balanced/Lenient), cost-vs-threshold analysis, and calibrated model artifacts.

    Python 5

  5. rag-qa-logs-corpus-data rag-qa-logs-corpus-data Public

    Synthetic multi-table RAG QA telemetry benchmark (corpus→chunks→retrieval→eval): labels for correctness/faithfulness/hallucination + cost/latency for RAG evaluation and dashboards.

    2

  6. llm-system-ops-production-telemetry-sft-data llm-system-ops-production-telemetry-sft-data Public

    Production-grade synthetic dataset for LLMOps: interaction-level telemetry (latency/cost/tokens), failure RCA, tool-use analytics, user feedback, plus 1:1 aligned SFT samples.

    1