Job description

Mission If you ask a hospital today how much a treatment will cost, the answer is usually: “We don’t know.” Not because providers don’t want to tell you but because patient cost is computed across insurance plans, negotiated rates, and billing rules that are fragmented, opaque, and not designed for real-time transparency. What should be a simple question requires navigating a web of systems that don’t talk to each other. So humans do that work instead. They log into portals, call payers, follow decision trees, and manually stitch together answers across disconnected systems. Even when providers want to give a clear answer, the system makes it nearly impossible. At Bravebird, we’re changing that. We’re building agents that do the work between systems - end to end. We believe the future of work is machines talking to machines, handling fragmented, system-to-system workflows so humans can focus on decisions, judgment, and care. 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. What You’ll Do • 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 What You’ll Bring • 3+ years of software engineering experience, 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 What We Offer • High ownership and impact • Competitive compensation with meaningful equity • Direct access to founders and customers • The opportunity to ship production systems in high-stakes environments We’re based in San Francisco and prefer working in person, but are flexible for exceptional remote candidates. Visa sponsorship available.

Requirements

Must-haves

If you ask a hospital today how much a treatment will cost, the answer is usually: “We don’t know.”
Not because providers don’t want to tell you but because patient cost is computed across insurance plans, negotiated rates, and billing rules that are fragmented, opaque, and not designed for real-time transparency.
What should be a simple question requires navigating a web of systems that don’t talk to each other.
So humans do that work instead.
They log into portals, call payers, follow decision trees, and manually stitch together answers across disconnected systems. Even when providers want to give a clear answer, the system makes it nearly impossible.
At Bravebird, we’re changing that.
We’re building agents that do the work between systems - end to end.
We believe the future of work is machines talking to machines, handling fragmented, system-to-system workflows so humans can focus on decisions, judgment, and care.
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.
What You’ll Do
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
What You’ll Bring

Nice-to-haves

Experience with fine-tuning or reinforcement learning
Experience building multimodal agents (voice, computer-use)
Experience in healthcare or other regulated environments

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimlllm

Platform: gem
Effort: medium