Note: The job is a remote job and is open to candidates in USA. PURE Insurance is a member-owned property and casualty insurer focused on providing exceptional service and smart insurance solutions. They are seeking a Senior AI Engineer to join their AI & Engineering team, where the role involves designing, building, and deploying secure and scalable AI solutions to improve operational performance and drive innovation.
Responsibilities
- Partner with the Lead AI Solutions Architect and AI Data Engineer to design, build, and deploy secure, scalable AI solutions: APIs, services, pipelines, agents, containers, and serverless functions that meet availability, performance, and security requirements. Deploy AI workloads primarily using cloud-native patterns, including AWS ECS-based containerized applications
- Build and operationalize LLM-enabled products including copilots, knowledge assistants, summarization engines, policy Q&A tools, and agentic workflows using Claude Code, OpenAI Codex, GitHub Copilot, AWS AgentCore Gateway, AWS AgentCore Harness, Databricks, and comparable LLM platforms. Apply thoughtful prompt and context patterns, tool/function calling, reusable agent skills, and agentic orchestration patterns
- Implement RAG, knowledge base, and document intelligence patterns end-to-end: ingestion, chunking, embeddings, vector and hybrid search, retrieval evaluation, and telemetry. Build and maintain Databricks-backed knowledge bases and AI agent capabilities where appropriate
- Deliver governed data and features for ML and GenAI — curated datasets, feature pipelines, and feature serving — supporting both training workflows and real-time inference with consistency, caching, backfill support, and latency SLOs
- Build reusable AI agent skills, tool definitions, prompts, guardrails, and orchestration patterns that can be shared across PURE’s AI products and engineering teams
- Use LangChain, LangGraph, or comparable frameworks where appropriate to build agent workflows, tool-use orchestration, stateful reasoning patterns, and multi-step automation
- Implement trust, safety, and governance controls including PII handling, prompt-injection defenses, content filtering, and policy-based access controls — built in close partnership with security and risk teams
- Ensure AI outputs are auditable, explainable, and compliant with applicable regulatory requirements (SOC 2, NAIC, GDPR) — a non-negotiable in insurance
- Define and maintain data lineage and model versioning practices so every production inference can be traced, reproduced, and reviewed
- Develop evals and red-teaming protocols to proactively identify failure modes in LLM-powered systems before they reach members
- Drive CI/CD, testing, versioning, reproducibility, and deployment standards across AI systems, including AWS ECS services, Databricks workflows, LLM applications, RAG pipelines, and agentic workflows
- Establish and maintain monitoring and observability across the full model lifecycle — from data ingestion through inference — including token/cost telemetry, latency dashboards, and drift detection
- Own incident response for AI platform issues: triage, root cause analysis, and remediation with appropriate urgency and communication
- Optimize cost and performance continuously: right-sizing compute, query tuning, caching strategies, and token budget management
- Support design and deployment readiness through architecture reviews, decision documentation (ADRs), and engineering standards that the broader team can build on
- Work across Engineering, Product, Data Science, Compliance, and business operations to translate member and business needs into AI-powered solutions
- Mentor engineers on the team; elevate technical quality through code reviews, pairing, and knowledge sharing
- Communicate clearly with both technical and non-technical stakeholders — able to explain an LLM tradeoff to an underwriter or a latency constraint to a product manager
- Stay current on the rapidly evolving AI landscape and bring actionable signal — new models, frameworks, patterns — back to the team
Skills
- 5+ years of professional software engineering experience, with at least 1 year building and operating AI/ML systems in production
- Proven hands-on experience with LLMs: prompt engineering, RAG pipelines, fine-tuning or adapting open-source models, function/tool calling, agent orchestration, and working with Claude, OpenAI/Codex, Gemini, or comparable models via API
- Experience building and shipping agentic AI systems, multi-step agents, tool-use orchestration, reusable agent skills, autonomous workflow automation, and governed enterprise integrations in a production environment. Experience with AWS AgentCore Gateway, AWS AgentCore Harness, LangChain, LangGraph, or comparable agent frameworks is highly valuable
- Strong Python engineering skills; ability to write clean, maintainable, production-grade code with FastAPI or similar frameworks, and package AI capabilities as APIs, services, workers, or containerized applications
- Experience with AI/ML infrastructure: Databricks-based AI/ML workflows, knowledge bases, vector or hybrid search, feature pipelines, model serving patterns, container orchestration, Docker, and cloud-native deployment
- Familiarity with cloud-native AI workloads on AWS — including cost governance and performance tuning at scale
- Experience implementing trust, safety, and governance controls in AI systems: PII handling, content filtering, access controls, and auditability
- Comfort working in a delivery-oriented team: you ship, you measure, you iterate
- Hands-on experience with AI-assisted software engineering tools such as Claude Code, OpenAI Codex, GitHub Copilot, or comparable developer productivity platforms
- Experience creating reusable AI agent artifacts such as skill files, tool definitions, prompt templates, system instructions, evaluation datasets, and guardrail patterns
- Experience with building reusable Github Workflow and make AI solutions part of the CI/CD
- Knowledge and experience with the Software Development Life Cycle (SDLC), including both low-code/no-code platforms and traditional application development using Java/Python
- Experience in building Serverless applications in AWS using AWS SAM
- Knowledge and experience with Terraform
- Background in insurance, fintech, or other regulated industries, with familiarity with compliance frameworks such as SOC 2, NAIC model laws, or GDPR
- Experience with LLM observability tooling: LangSmith, Weights & Biases, Dynatrace LLM monitoring, OpenTelemetry, or equivalent
- Familiarity with ML frameworks (PyTorch, scikit-learn) for classical ML alongside LLM-based approaches
- Experience with fraud detection, document intelligence, risk scoring, or actuarial data systems
- Contributions to open-source AI/ML projects or published work in applied NLP or machine learning
- Experience deploying AI applications as containerized services on AWS ECS or similar cloud-native platforms
- Experience designing and implementing agent skills, tool registries, function-calling interfaces, or reusable agent capabilities for enterprise AI systems
Company Overview