Forward Deployed AI Engineer

Forward Deployed AI Engineer

$70 - $85 per hour | Exeter, RI | Remote | Consulting

Remote contract-to-hire that converts to $225K. Put production agentic AI into a mission-driven health plan, own the platform from day one, and help a team grow from 50 to 150 engineers. Your first-week onsite costs are covered.

A bit about us:

We are a national, AI-first technology solutions provider. We help enterprise and public-sector organizations modernize how their people work, through AI, cloud, collaboration and managed services. We’ve been nationally recognized as a best place to work for career growth. We embed senior engineers inside our clients’ teams to turn AI strategy into production systems that last. Our current engagement is with a regional, not-for-profit health plan that has already shipped AI to production on its own and is now insourcing engineering as a deliberate strategy, growing its team from about 50 to 150. As a Forward Deployed AI Engineer, you will stand up the plan’s enterprise AI platform, build the guardrails and reference patterns everything else is built on, and then go deep on workflows where success is measured in members served.

Why join us?

• Contract-to-hire, converting to a full-time role with us at a $225,000 base salary
• Fully remote after an expense-paid first week onsite
• Ground-floor ownership: you build the eval framework, the guardrails and the orchestration patterns
• Real production problems in prior authorization, appeals and member services, not demos
• An employer recognized nationally for career growth
• Full benefits package upon conversion

Job Details

What We Are Looking For:

This isn’t a seat where you build demos for a slide deck. We are looking for a backend/platform engineer who has personally taken LLM-backed or agentic workflows past proof of concept into production, and owned them afterward through evals, guardrails, cost control and observability. You’ll be embedded with the engineering team of a regional health plan in a PHI-everywhere environment. You will run discovery with business stakeholders and pair hands-on with engineers to deliver. The ideal candidate is equally comfortable defining the right problem with a non-technical business owner and writing the code to solve it. Just as important, they know when a problem doesn’t need AI at all. This client describes itself as a “Toyota Camry shop”: reliable, maintainable and simple beats sophisticated. The team will also be heavily involved in designing and building a Care Management System, where compliance with NCQA standards is critical. This role is remote, with minimal travel to Providence, RI, and requires onsite attendance the full week of 11/9.

What You Will Do:

• Stand up Claude Code and Azure AI Foundry for a 20-seat developer pilot, including Entra authentication, quota management and cost controls.
• Build guardrails, secure design patterns and reference implementations that the client’s engineering teams build on.
• Author and secure MCP servers and tool integrations connecting agents to ServiceNow, Snowflake, Grafana/OpenTelemetry, Entra and healthcare data systems.
• Build the AI eval and test harness covering prompts, retrieval, tool calls, permissions and end-to-end workflows.
• Design and deliver multi-step and agentic workflows for appeals and grievances, prior authorization and member services. Pick the right shape for each: deterministic workflow, single agent or multi-agent.
• Build RAG pipelines over the policy corpus with access controls, hybrid retrieval, citations and grounding checks.
• Implement policy-as-code and automated audit evidence for an organization audited seven times a year.
• Lead discovery sessions with business owners, document current-state workflows, and translate needs into requirements, user stories and measurable success criteria.
• Instrument AI workloads for observability, including traces, token and cost telemetry, and SLOs.
• Pair with client engineers and transfer knowledge so what you build stays supportable without you.

What We Provide:

• Contract-to-Hire (6 months)
• $225,000 base salary upon conversion
• Remote, with expense-paid onsite sessions
• Full Benefits Package Available upon conversion

What We Believe:

We are committed to creating a diverse and inclusive environment and are proud to be an equal opportunity employer. We do not and will not discriminate in employment and personnel practices on the basis of race, sex, age, disability, religion, national origin, or any other basis prohibited by applicable law. Hiring, transferring, and promotion practices are performed without regard to the above listed items.

Requirements

What We Require:

• U.S. Citizenship required
• 6-10+ years designing, building, deploying and supporting production applications in enterprise or regulated environments, with a backend/platform focus
• At least one LLM-backed or agentic workflow shipped to production with real users and owned after launch, with the ability to speak to what broke and how you fixed it
• Strong Python and working TypeScript
• Hands-on experience building tool integrations or MCP servers, and an eval harness for an LLM application
• Hands-on RAG experience: ingestion, chunking, embeddings, hybrid search, reranking, access-controlled retrieval and retrieval evaluation
• Azure fluency: AI Foundry, Entra ID, Data Factory, AI Search, AKS/Container Apps, Key Vault and Azure DevOps
• Working knowledge of Snowflake
• Experience with regulated data (healthcare/HIPAA strongly preferred, financial services acceptable)
• Proven business-analysis skills: leading discovery, defining requirements and acceptance criteria, and explaining trade-offs to executives
• Ability to clearly explain the cost story and eval story behind what you’ve built
• Healthcare payer domain knowledge (claims, eligibility, prior authorization, appeals, care management, NCQA, Stars) strongly preferred
• Experience with Claude Code and the Claude Agent SDK preferred