Hey, I'm Erik Cohen 👋

Senior AI Engineer | Evaluation-First AI Systems for Healthcare & Life Sciences

About Me

I build AI systems that actually work in production, not just demos. My specialty is bridging the gap between impressive prototypes and reliable, measurable systems that scientists and researchers depend on.

Currently running Cohen AI Consulting LLC, helping organizations build healthcare AI systems with evaluation-first methodology. Previously spent 4 years at BenchSci where I reduced manual validation time by 60% with GenAI evaluation pipelines, scaling the platform to 40+ enterprise pharma clients.

US Army veteran. Pursuing my MS in Computer Science (Machine Learning) at Georgia Tech. When I'm not building AI systems, I'm in the kitchen experimenting with recipes or hanging out with my daughter.

What I'm Focused On

• Autonomous AI systems that can reason and achieve goals
• Evaluation frameworks for non-deterministic AI
• Healthcare and life sciences AI applications
• Multi-agent orchestration patterns

Tech Stack

• AI/ML:

• Full Stack:

• Infrastructure:

Career Highlights

🏥 Cohen AI Consulting (2025-Present)

• Independent AI consulting for healthcare and life sciences
• Evaluation-first methodology: quality gates, observability, confidence metrics
• Helping organizations move from AI demos to production systems that work

🧬 BenchSci — Senior AI Engineer (2021-2025)

• Reduced validation time 60% with GenAI evaluation pipelines
• Scaled platform to 40+ enterprise pharma clients
• Built AI tools helping scientists design better experiments

🌍 National Geographic — Technical Lead (2017-2019)

• Led YourShot community platform serving millions of monthly users

🤖 Primer.ai — Full Stack Engineer (2021)

• Built ML-powered risk analysis tools for intelligence applications

Education

🎓 Georgia Tech — MS Computer Science, Machine Learning (Expected 2026)

Let's Connect

Building AI for healthcare or life sciences? Need help turning demos into production systems? Let's talk.