# Waleed Barakat

> AI Engineer & Founder, building data and AI systems that ship to production at scale.

- **Location:** USA
- **Email:** wbarakat94@gmail.com
- **GitHub:** https://github.com/wbarakat
- **X:** https://x.com/WaleedB999
- **Schedule a call:** https://calendly.com/wbarakat94/30min

## About

I build data and AI systems that ship to production at scale. Co-founder of [bzar.app](https://bzar.app), previously Senior Data Scientist at [Strive Health](https://www.strivehealth.com/).

## Experience

- **Co-Founder · bzar.app** (2025 – Present, Remote)
  Owning product, data, and the AI stack end to end, from prototype to production.

- **Senior Data Scientist · Strive Health** (Sep 2024 – Present, Remote)
  Trained XGBoost and AutoML models that predict 30-day hospitalization risk for 120k CKD patients, driving an 18% drop in unplanned admissions. Built an end-to-end MLOps pipeline (dbt feature marts in Redshift → SageMaker → Lambda scoring), shrinking model retrain time from 2 weeks to 2 hours.

- **Data Engineer · USDA** (Aug 2023 – Sep 2024, Remote)
  Optimized 30+ ETL pipelines in Python, SQL, and cron, cutting data processing time 30% while holding 99.9% uptime across critical systems. Automated Tableau reporting, halving manual report generation time.

- **Senior Data Scientist · Appian** (Aug 2022 – Jul 2024, McLean, VA)
  Deployed a churn-prediction model on GCP (BigQuery + PostgreSQL) that drove a $10M reduction in customer churn, surfaced to execs through live Tableau dashboards.

- **Data Scientist · Henry M. Jackson Foundation** (Dec 2019 – Aug 2022, Annandale, VA)
  Built mass-spec and RNA-Seq pipelines in R (+40% processing efficiency) and RShiny QC dashboards (−25% instrument downtime), contributing to four peer-reviewed publications.

## Publications

- [Histology-resolved proteomics reveals distinct tumor and stromal profiles in low- and high-grade prostate cancer](https://doi.org/10.1186/s12014-025-09534-8) · Clinical Proteomics, 2025
- [Mapping three-dimensional intratumor proteomic heterogeneity in uterine serous carcinoma by multiregion microsampling](https://doi.org/10.1186/s12014-024-09451-2) · Clinical Proteomics, 2024
- [In situ profiling reveals metabolic alterations in the tumor microenvironment of ovarian cancer after chemotherapy](https://doi.org/10.1038/s41698-023-00454-0) · npj Precision Oncology, 2023
- [Brain proteomic atlas of alcohol use disorder in adult males](https://doi.org/10.1038/s41398-023-02605-0) · Translational Psychiatry, 2023
- [Classification of high-grade serous ovarian cancer using tumor morphologic characteristics](https://doi.org/10.1001/jamanetworkopen.2022.36626) · JAMA Network Open, 2022
- [Extensive three-dimensional intratumor proteomic heterogeneity revealed by multiregion sampling in high-grade serous ovarian tumor specimens](https://doi.org/10.1016/j.isci.2021.102757) · iScience, 2021

## Featured Projects

- **[Loadout](https://github.com/wbarakat/loadout)**: Local-first, E2E-encrypted vault for your agent setup: skills, memory, and secrets synced across Claude Code, Cursor, Codex, and every agent tool you use.
- **[ASRE](https://asre-docs.vercel.app)** ([repo](https://github.com/wbarakat/asre)): Admission Signal Reliability Engine, a healthcare data reliability layer that stitches ADT, claims, authorization, and eligibility signals into one trusted, deduplicated admission events table with confidence scores.

## Tech Stack

Python, R, SQL/PostgreSQL, pandas, NumPy, scikit-learn, PyTorch, LangChain, LangGraph, Hugging Face, Anthropic/Claude, Cursor, GitHub Copilot, Ollama, Airflow, Spark, BigQuery, Snowflake, Jupyter, Docker, Terraform, Git, PostHog, Sentry, Stripe, Supabase, Vercel, Clerk, Linear, Redis, Resend.

## GitHub Activity

Real-time contribution graph (public + private via authenticated API): https://waleedbarakat.dev, account [`wbarakat`](https://github.com/wbarakat).