About the role
We're building agentic AI systems that can reason, act, and operate reliably in messy, real-world environments - across chat, voice, and full computer-use interfaces. As an Applied ML engineer, you'll help build the core intelligence powering these systems: reasoning, planning, grounding, memory, evaluation, and reliability. This is a deeply technical, high-ownership role. You'll work directly with founders, shape the technical direction, and ship systems that are used in real-world, high-stakes environments.
Responsibilities:
- Design and implement reasoning and planning systems for real production workflows.
- Build robust tool orchestration, grounding, and execution frameworks.
- Develop memory, context, and state management for long-horizon tasks.
- Create evaluation systems that measure correctness, reliability, and performance.
- Work across chat, voice, and UI-based agents to make them coordinated and dependable.
- Translate research ideas into production systems.
- Partner closely with founders and customers to iterate quickly and ship.
- Help define engineering culture, standards, and early technical direction.
Requirements:
- 5+ years of software engineering experience.
- 2+ years building and shipping LLM or agentic systems in production.
- Experience with evals, memory, retrieval, and grounding architectures.
- Comfort operating in ambiguous, open-ended problem spaces.
- Strong bias toward building reliable, measurable, testable systems.
- Product intuition - you care about real users and real outcomes.
- Clear communication and strong collaboration skills.
Bonus:
- Experience with fine-tuning or reinforcement learning.
- Experience building multimodal agents (voice, computer-use).
- Experience in healthcare or other regulated environments.