Agentic Lifecycle Engineer

Enabuild AI; Boston (MA); Hybrid; USD 140,000 - 190,000; Full time

Benefits offered by this job

Remote work

Hybrid/Onsite options

Job summary

Enabuild AI is seeking an Agentic Lifecycle Engineer to own the full lifecycle of enterprise agents. You will design and operate the runtime layer including harnesses, tools, and memory policies, while driving adoption and governance across data, people, process and policy.

Role supports remote US or hybrid/on-site options in Westwood, MA, with a base salary plus bonus depending on experience. You will report to the President of Enabuild AI and contribute to a responsible, scalable AI estate.

Qualifications

• Hands-on experience building or operating LLM agents beyond prototypes.
• Strong Python and/or TypeScript, APIs, distributed systems.
• Prompt/context/harness engineering and RAG/GraphRAG.
• Testing/observability mindset with evals and telemetry.
• Cloud security fundamentals and least privilege access.
• Ability to explain trade-offs to engineers and CFOs.

Responsibilities

• Own the runtime layer that decides success: harness, tools, context, interfaces.
• Assess readiness and define operating model before build starts.
• Design and build the agent and Above the Model layer.
• Drive change management, governance, and post go-live operations.
• Run agent estate: drift detection, performance, evolution and FinOps.

Skills

LLM agents

Python/TypeScript

APIs

Harness engineering

RAG GraphRAG

Tool calling

Observability

Cloud security

Tools

MCP

Job description

ENABLE THE WORK. BUILD THE AGENT. OWN THE LIFECYCLE.

What You’ll Own

ENABLE - Assess readiness (data, people, process, policy, platform). Define operating model and a forward-declared business value commitment before build starts.

BUILD - Design and build the agent plus the Above the Model layer: harness, tools, context, APIs/MCP, RAG/GraphRAG, evals, sandboxes, digital twins.

ADOPT - Drive change management, human-in-the-loop design, feedback loops, and the 4 Abilities (Visibility, Observability, Traceability, Accountability). Make governance operational.

ASSURE (AIOMS) - Run the agent estate post go-live: drift detection, performance, evolution, skills/data management, and continuous FinOps.

Six foundations sit under every stage: Data People Process Policies Value Ecosystem.

Above the Model (where you work)

You own the runtime layer that decides if an enterprise agent succeeds: harness (retries, budgets, circuit breakers, kill switch), tools (least-privilege, versioning), context/memory policy, interfaces (API/MCP), evals, orchestration, and instrumentation (traces, cost, drift). Models change; this layer is the durable IP.

What You Bring

• Hands-on experience building or operating LLM agents beyond prototypes
• Strong Python and/or TypeScript, APIs, distributed systems
• Prompt/context/harness engineering, RAG/GraphRAG, tool calling, agent frameworks (MCP a plus)
• Testing/observability mindset (evals, traces, telemetry)
• Cloud + security fundamentals (identity, least privilege, secrets)
• Ability to explain trade-offs to both engineers and CFOs
• Ownership mindset — you care about post go-live as much as pre

Nice-to-haves: multi-agent orchestration, LLMOps/AIOps/SRE, FinOps/token economics, change management, AI governance (NIST, ISO 42001, EU AI Act), knowledge graphs, synthetic data, digital twins, enterprise integration.

Why Enabuild AI

Own the full lifecycle. Shape reusable IP (harnesses, tools, evals) from the ground floor. Structured cross-training across all stages. Work where the real 80% lives — above the model. Responsible AI that amplifies people, with governance and FinOps built in.

Details

• Role: Agentic Lifecycle Engineer (all four tracks)
• Location: Remote (US) / Hybrid or Onsite Westwood, MA
• Full-time | Base + bonus, commensurate with experience
• Reports to: President of Enabuild AI

Enabuild AI is an equal opportunity employer.

ENABLE THE WORK. BUILD THE AGENT. OWN THE LIFECYCLE.