Mohcine Madkour

Senior AI/ML Engineer & Architect · PhD in Computer Science · Plano, TX
GitHub · LinkedIn · X

prof_pic.jpg

📍 Plano, TX (DFW Area)

🎓 PhD · Computer Science

🏥 Postdoc · Biomedical Informatics

Available for hybrid/remote roles

I build and ship production AI • AI/ML Engineer & Architect • PhD Computer Science / Postdoc Biomedical Informatics, University of Texas Health Science at Houston • Previously: healthcare AI, surgical robotics, industrial IoT

Currently an AI/ML Engineer and AI Architect consultant focused on agentic AI systems, RAG evaluation, and MLOps at scale. I studied Computer Science and completed postdoctoral work in Biomedical Informatics at the University of Texas Health Science at Houston. Previously built predictive maintenance MLOps for surgical robot fleets at Intuitive Surgical, connected diagnostics at Cummins, and co-developed the MySurgeryRisk clinical AI platform at UTHealth — powering surgery risk prediction across thousands of patients.


What I Build

I specialize in the full ML delivery lifecycle — from prototype to production:

  • Agentic AI Systems — LangGraph orchestration, multi-agent pipelines, RAG architectures with ChromaDB and LangSmith evaluation
  • MLOps & CI/CD — MLflow tracking, automated model monitoring (Evidently, Arize Phoenix, Azure ML Monitor), GitHub Actions pipelines, CT/CD for ML
  • Healthcare AI — HL7/FHIR integration (Mirth Connect certified), clinical risk prediction (MySurgeryRisk, AUC 0.82–0.94), HIPAA-compliant data pipelines
  • Computer Vision — YOLOv8 + ByteTrack for real-time object detection and tracking (surgical robotics, warehouse systems)
  • Cloud & Infrastructure — Azure ML, AWS SageMaker, Databricks, PySpark, GCP

Recent Impact

  • 📉 25% reduction in unplanned downtime for a global surgical robot fleet (Intuitive Surgical) — built predictive maintenance system with MLOps CI/CD and $700K+ savings
  • 🏥 MySurgeryRisk at UF Shands Hospital — clinical AI platform powering surgery risk prediction (AUC 0.82–0.94) with HL7 FHIR pipelines and Mirth Connect
  • 🏭 Cummins PreventTech — IoT predictive diagnostics at scale using PySpark, Azure ML, and Databricks for connected engine fleets

Core Stack

LangChain · LangGraph · LangSmith · ChromaDB · MLflow · PyTorch · YOLOv8 · ByteTrack · Azure ML · Databricks · AWS SageMaker · PySpark · Streamlit · FastAPI · Python · Node.js


Currently

🎓 Building a Udemy course: “Agentic AI Evaluation: Production-Ready Systems with LangGraph and LangSmith”

📝 Writing about agentic AI systems, RAG evaluation, and production ML on this site

🔬 Assistant Research Professor · Published in biomedical informatics and clinical AI

news

Apr 01, 2026 🎓 Launching Udemy course: “Agentic AI Evaluation: Production-Ready Systems with LangGraph and LangSmith” — M1 & M2 recording in progress.
Mar 31, 2026 🎤 Spoke at Convergence AI Dallas (Irving Convention Center) on agentic AI in production — evaluation, monitoring, and the gap between demos and reliable systems.
Jan 15, 2026 🔬 Open to Senior AI/ML Engineer, AI Architect, and Head of AI roles in DFW (hybrid) and remote. Reach out via LinkedIn or email.

latest posts