AI Engineer

What you'll do

You will architect and implement the agentic layer of our core product. This involves creating sophisticated AI systems capable of reasoning, executing actions, and reliably managing complex workflows across critical business functions such as pricing and underwriting, policy issuance, claims intake and adjudication, fulfillment (repair/replacement/reimbursement), and other operational areas.

Key responsibilities

• Design, develop, and deploy production-grade AI agents to automate real business processes, moving beyond simple demonstrations.
• Construct robust agentic architectures encompassing orchestration, tool integration, state management, memory, permissions, audit trails, human-in-the-loop mechanisms, and fallback strategies.
• Take end-to-end ownership of our Retrieval Augmented Generation (RAG) platform, including data ingestion, chunking, embedding generation, retrieval, reranking, citation/grounding, and hallucination mitigation techniques.
• Build sophisticated evaluation and monitoring systems, including offline evaluation datasets, regression tests, online performance metrics, drift detection mechanisms, and red-teaming suites.
• Implement advanced model optimization techniques, such as prompt engineering, structured output generation, strategic fine-tuning, latency and cost optimization, caching, and throughput tuning.
• Develop core Machine Learning systems for warranty and claims processing, focusing on document understanding, data extraction, classification, anomaly and fraud detection, decision support, and service level agreement (SLA) routing.
• Collaborate closely with product and operations teams to translate intricate real-world workflows into deterministic, testable, and compliant automated solutions.

What you'll build (examples)

• Underwriting/pricing agents: Develop agents for real-time quote decisions, incorporating merchant, product, and contextual signals with strict adherence to guardrails and auditability requirements.
• Claims copilot + auto-adjudication engine: Create systems for claims intake triage, evidence request management, decision proposals with comprehensive explanations, vendor routing, and automated reimbursement processes.
• OEM warranty parsing system: Build solutions to transform unstructured manufacturer warranty policies into machine-readable coverage logic.
• Internal ops copilots: Develop tooling designed to reduce manual effort and enhance consistency across customer support, compliance, and finance departments.

Requirements (must have)

Note: Hiring at various levels. Required experience and skill level will adapt to the specific role level.

• Minimum of 1 year of experience building and deploying ML/LLM systems in a production environment, or equivalent founder-level experience.
• Demonstrated experience in building agentic products or companies, including expertise in multi-step workflows, tool utilization, orchestration, and reliability engineering.
• Deep hands-on expertise in:
  • RAG and retrieval systems, including vector databases, reranking strategies, and grounding techniques.
  • LLM evaluation methodologies, such as golden sets, automated judging, human evaluation, and regression pipelines.
  • Prompt engineering and structured output generation, including schema design, function/tool calling, and robustness.
  • Fundamentals and trade-offs of model training and fine-tuning (understanding when to tune versus prompt or retrieve).
• Strong software engineering skills, including designing clean APIs, implementing comprehensive testing strategies, ensuring observability, performance tuning, and adopting a secure-by-default design philosophy.
• Comfort and proven ability in owning ambiguous problems end-to-end and driving them towards measurable, positive outcomes.

Strong preference (nice to have)

• Experience building systems subject to compliance and audit requirements, particularly within fintech, insurance, healthcare, or enterprise sectors.
• Experience with large-scale document AI applications, including handling diverse formats (PDFs, images, messy inputs) and reliably extracting structured information.
• Experience designing human-in-the-loop workflows and sophisticated escalation rules for critical decision-making processes.
• Experience with infrastructure for LLMs, such as model hosting, batching, streaming, caching, and prompt/version management.
• A background in startups or ex-founder experience, especially in rapidly shipping 0-to-1 products.

What success looks like (first 90 days)

• Successful deployment of an agentic workflow that significantly reduces manual operations work and demonstrably improves a key metric (e.g., cycle time, accuracy, cost per claim, attach rate, CSAT).
• Implementation of a robust evaluation harness capable of identifying regressions before they reach production and providing a reliable quality score for each workflow.
• Establishment of a scalable architectural pattern for agents, covering aspects like permissions, audit logs, observability, and fallback mechanisms, which can be adopted and replicated by the team.

Tech environment

Our technology stack is cloud-native and designed for rapid iteration. We primarily use Python for ML and agent development, TypeScript for user-facing product interfaces, and PostgreSQL for our systems of record. Our architecture is event-driven, featuring a modern LLM and retrieval stack augmented by strong observability tools and CI/CD practices. Infrastructure is managed across AWS and Azure.

Why this role is special

• Opportunity to build an AI-native, category-defining company within a vast and impactful market.
• Direct exposure to founders and high leverage: your contributions will significantly influence the company's strategic direction.
• Exceptional breadth of experience, covering growth initiatives, underwriting and claims operations, and product development, all within a single role.
• A significant career accelerant: demonstrating strong performance will lead to rapid growth in scope and title.

How To Apply

• Ensure your profile is current and includes a link to your LinkedIn profile.
• In your application message, please provide three concise sentences, each describing something you have built or delivered along with the results achieved.