Forward Deployed AI Engineer

Amigo•New York, NY

1d•$130,000 - $160,000

About The Position

About Amigo Amigo partners with healthcare organizations to deploy robust AI infrastructure that directly serves patients and providers. Our agents handle clinical workflows and patient engagement across the entire journey: pre-visit intake, care navigation, post-visit care plans, patient monitoring, and more. We're fresh off our Series A backed by Tier 1 investors like Madrona, General Catalyst, and Optum Ventures. Our work is validated with leading academic medical institutions. Our agents have reached 3M+ patient encounters and are on track to 10x this year. About this role The Applied AI Engineer role at Amigo is a launchpad. You'll ship production AI agents end-to-end — writing the code, shaping product decisions, and sitting with customers as they deploy — and in doing so, you'll build the exact skill set that opens doors to the next chapter of your career at Amigo. This role is designed to help you grow into other parts of the company. Applied AI Engineers move into one of four paths within 6–12 months: Applied AI: stay in the org; go deeper on agent architecture, verification frameworks, and the hardest customer problems; become a technical leader on the applied team Product Management: if you gravitate toward roadmap, discovery, and shaping what we build next Sales Engineering: if you thrive in pre-sales, technical storytelling, and closing enterprise deals Software Engineering: if you want to go deeper on platform, infrastructure, and core systems You won't have to guess which fits. This role will actively help you explore each path, give you projects that stretch in those directions, and sponsor the internal transition when you're ready. People who want to stay on the applied side long-term can do that too — but this role is explicitly built so you don't have to.

Requirements

2+ years of production software engineering experience
Strong Python skills including proper use of typing, and experience building highly reliable systems that interact with external APIs (schema design, retry strategies, error propagation)
Experience with LLMs, prompt engineering, and building on AI platforms
Ability to implement systems with strict reliability requirements
Understanding of testing and verification methodologies
Experience working directly with customers or stakeholders to deliver technical solutions
Familiarity with git workflows and collaborative development
Strong debugging and problem-solving skills for complex systems
Ability to balance theoretical concepts with practical implementation constraints
Clear technical communication for both engineering and non-technical audiences

Nice To Haves

Experience in regulated industries (healthcare, finance, legal)
Background with simulation, synthetic data, or evaluation frameworks
Understanding of distributed systems and service-oriented architectures
Experience with observability and monitoring in production environments
Familiarity with compliance requirements (HIPAA, SOC 2, GDPR)

Responsibilities

Implement context graphs that enable agents to select the right reasoning mode (lookup, pattern recognition, exploration) based on problem complexity
Build and optimize agent configurations, including static personas, dynamic behaviors, and functional memory systems
Design bounded Operational Patient Domains (OPDs) with explicit inclusions, exclusions, capabilities, and escalation rules
Implement verification loops and simulation-based testing to validate agent performance against customer KPIs
Write Python code for tool integrations and API orchestration within customer environments
Build evaluation suites that gate deployments using customer-specific success metrics
Collaborate with domain experts through structured interviews to capture reasoning patterns and clinical protocols
Support technical power users through git-based workflows for rapid iteration
Implement adversarial testing to systematically identify and prevent failure modes
Ensure agents maintain audit-ready provenance with version pinning and evidence links
Contribute to systematizing implementation patterns across different problem domains