AI/ML Technical Leader

Moving ML from proof of concept to production

I help organizations build AI systems that actually ship. For the past 8 years, I've specialized in engineering ML for safety-critical applications—DoD edge deployments, healthcare conversation intelligence, IoT monitoring systems—building systems your entire team can maintain, not just ML specialists.

Jay Palat

Proven Impact

Results that matter

2wk → 4hr

Retraining Time

Reduced ML model retraining cycle from 2 weeks to 4 hours for DoD edge-deployed computer vision threat detection system.

10→10+/day

Patient Throughput

Scaled mental health assessment platform from 10 patients per month to 10+ per day while reducing evaluation time from weeks to minutes.

+15%

Model Accuracy

Improved model accuracy while cutting ML pipeline delivery time from weeks to days through systematic data quality improvements.

Services

What I do

MLOps & Production AI

Design and deploy ML pipelines that actually ship. Edge computing, model serving, continuous training, and monitoring at scale.

Team Building & Leadership

Rebuild engineering organizations, hire technical leadership, and establish processes that deliver high-quality software products.

Safety-Critical AI

Engineering AI for defense, healthcare, and other environments where reliability isn't optional. Security clearances held.

Fractional AI Leadership

Part-time Head of AI/ML for organizations at the inflection point between prototype and production-ready systems.

Background

Experience

2023 – 2025

Head of AI/ML

mpathic

Led ML platform development for real-time conversation analytics in healthcare and life sciences. Built Transformer-based text classification pipelines, designed OpenAI API applications for zero-shot behavior identification, and established AI organizational practices.

2018 – 2022

Technical Director, AI for Mission

Software Engineering Institute, Carnegie Mellon University

Led strategic direction for AI Engineering in mission-critical DoD applications. Managed 7 managers and 40 direct reports. Built edge AI pipeline for computer vision threat detection deployed with US Army. Held TS-SCI clearance.

2016 – Present

Principal

QOE Labs

Fractional technical leadership for startups and established companies including BCG, CMU, and multiple healthcare/IoT ventures. Rebuild engineering teams, architect systems, and advise on AI/ML strategy.

2014 – 2016

Sr Director of Data & Operations

Rhiza (acquired by Nielsen)

Led technical transformation from prototype to acquisition-ready product. Implemented continuous deployment, managed cross-functional teams of 9-15 engineers.

2000 – 2010

Certified Consulting IT Specialist

IBM

Technical lead for enterprise web applications including IBM.com Support Portal. Led architecture teams of 3-30 across multiple locations. Developed SOA governance and 3rd-party integration frameworks.

Thought Leadership

Publications & Patents

Let's talk

Working on ML in production, building technical teams, or deploying AI in safety-critical systems? I'm always happy to discuss interesting problems.