Director, Foundational AI & World Models

Merlin; Boston (MA); Hybrid; USD 180,000 - 280,000; Full time

Benefits offered by this job

Remote work possible

Hybrid schedule in Boston HQ

Job summary

Merlin Labs in Boston, MA is seeking an experienced AI leader to own Merlin's foundation and world-model work, directing architecture, build-versus-adapt choices, and the roadmap for autonomy AI components. You will lead and mentor a small team of world-model engineers, define evaluation suites, and ensure safety-conscious outputs and regulator-ready design across the autonomy stack.

You will collaborate with Systems Engineering, Certification, and the Chief Architect to keep designs defensible

Qualifications

• 8+ years building AI systems and 3+ years leading technical teams or owning a major model program.
• Proven experience shipping AI models into production in real-time environments.
• Deep knowledge of world models, planning, and evaluation practices.
• Strong PyTorch background and ability to reason about real-time systems in C++.

Responsibilities

• Technical strategy: own foundation and world-model work and define the capability roadmap.
• Team leadership: mentor a small team of world-model and post-training engineers.
• System interface: design model interfaces to the autonomy stack with schema-constrained outputs.
• Evaluation: build evaluation harnesses and regression suites for model releases.
• Safety outputs: implement uncertainty quantification and out-of-distribution detection.
• Benchmarking: compare learned planning with rule-based behavior across missions.
• Certification collaboration: work with Systems Engineering and Certification to stay defenisible.
• Research judgment: monitor external advances and make disciplined integration decisions.

Skills

AI systems

Team leadership

Production deployment

PyTorch

C++

Architecture decisions

Safety/regulatory mindset

Education

CS/AI/related degree

Tools

PyTorch

C++

Job description

Merlin Labs in Boston, MA is seeking an experienced AI leader to own Merlin's foundation and world-model work, directing architecture, build-versus-adapt choices, and the roadmap for autonomy AI components. You will lead and mentor a small team of world-model engineers, define evaluation suites, and ensure safety-conscious outputs and regulator-ready design across the autonomy stack.

You will collaborate with Systems Engineering, Certification, and the Chief Architect to keep designs defensible