ML Engineer - Automatic Computer Use (Windows)

What You Will Do

• Design, train, and deploy ML models for computer usage and automated task execution, including UI understanding, system control, and workflow automation

• Build and scale Large Action Models (LAMs) that translate user intent into structured plans, tool calls, and executable actions

• Develop agentic AI pipelines that combine perception (screen, UI, system state), reasoning, planning, and execution with strong reliability guarantees

• Implement and evaluate action execution frameworks, including tool orchestration, sandboxed execution, rollback, and failure recovery

• Train and fine-tune multimodal models (LLMs, VLMs) for screen understanding, element grounding, and instruction-following

• Partner with OS, application, and platform teams to integrate AI agents deeply into Windows and cross-device environments

• Build evaluation systems for task success, robustness, latency, and safety in real-world computer usage scenarios

• Collaborate with product, UX, and security teams to ensure safe, transparent, and controllable AI behavior

• Stay at the forefront of agentic AI, tool-using LLMs, large action models, and autonomous execution systems

Basic Qualifications:

• At least 5 years of experience in ML engineering, applied AI, or intelligent systems

• Bachelor's Degree in Computer Science, Engineering, or a related technical discipline

• Strong experience with LLMs and/or multimodal models for reasoning, planning, or control

• Hands-on experience building end-to-end ML systems, not just models (training evaluation deployment)

• Proficiency in Python and experience with modern ML frameworks (PyTorch, JAX, TensorFlow)

• Experience shipping ML-powered features into production products

Preferred Qualifications:

• Experience working on computer usage agents, UI understanding, or automated workflow execution

• Familiarity with Large Action Models (LAMs), tool-using agents, function calling, or planner-executor architectures

• Experience with multimodal learning, including vision + language grounding

• Knowledge of reinforcement learning, imitation learning, or human-in-the-loop training

• Experience deploying ML systems on Windows, Android, or edge devices

• Strong intuition for failure modes, safety, and reliability in autonomous systems

• Passion for building AI that actually does things, not just talks about them