You’ll work shoulder-to-shoulder with customer teams , co-owning delivery from prototype → production . This role is ideal for an engineer who can build full-stack software , thinks like a product owner, and can translate ambiguous business workflows into reliable, scalable AI agent systems on Lyzr.

You will also serve as a feedback conduit—bringing what you learn in real deployments back into Lyzr’s product and engineering roadmap.

What you’ll do

• Co-architect and co-build production-grade AI agent applications with enterprise customers—on-site—covering everything from initial scoping to production rollout.
• Lead technical discovery and design sessions : clarify requirements, map workflows, define success criteria, and propose an implementation plan that balances speed, reliability, and maintainability.
• Build the full application layer :
• Front-end experiences for agent workflows (e.g., dashboards, review/approval flows, human-in-the-loop)
• Backend services and APIs
• Data + database layer (relational + vector + metadata stores)
• The agent layer on Lyzr (orchestration, tools, memory/context, retrieval, evals)
• Implement agentic AI best practices : prompt engineering, context engineering, tool calling, guardrails, evaluation, observability, and cost/performance optimization.
• Deliver robust Graph + Graph RAG systems when customer problems demand multi-step reasoning and structured retrieval.
• Partner with Sales / Solutions / GTM to define implementation strategies, unblock customer success, and accelerate time-to-value.
• Identify patterns, friction points, and missing capabilities from deployments; synthesize learnings into actionable product feedback for Lyzr’s roadmap.
• Stay current on the agentic AI ecosystem and translate new capabilities into practical, secure, customer-ready solutions.

What we’re looking for****Must-haves

• Strong customer-facing engineering ability: you can run technical sessions, manage stakeholders, and communicate tradeoffs clearly (technical + non-technical).
• Proven experience building full-stack software that ships:
• Front-end (React/Next.js or similar)
• Backend services (Python/FastAPI, Node/TypeScript, or equivalent)
• Databases (Postgres/MySQL) and experience with data modeling
• Hands-on experience building LLM/agentic applications in production or near-production environments, including:
• Agent orchestration, tool calling
• Prompt + context engineering
• RAG patterns; Graph RAG a strong plus
• Strong engineering fundamentals: code quality, testing, debugging, performance, reliability, and scaling.
• Ability to operate independently, make good judgement calls, and keep customers/internal teams aligned via crisp updates and deliverables.
• Willingness to be primarily onsite with customers in NYC (this role is designed to be forward‑deployed).

Nice-to-haves (highly valued)

• Prior experience with LangChain / LangGraph , CrewAI , or similar frameworks (shows transferable agent-building intuition).
• Experience with MCP and APA integrations , and designing tool/plugin ecosystems.
• Production deployment experience with Docker/Kubernetes , CI/CD, and cloud infrastructure (AWS/GCP/Azure).
• Experience implementing enterprise concerns: auth (SSO/SAML/OAuth), audit logs, permissions, multi-tenancy, data governance, PII handling.
• Former founders or engineers with “unusual paths” who have shipped real products and can thrive in ambiguity.

Example projects you might work on

• Multi-agent enterprise workflows (e.g., support triage, analyst copilots, compliance review, sales ops automation)
• Graph-based agent systems that coordinate retrieval, reasoning, and tool use across multiple sources
• Secure, scalable internal AI apps with approvals, traceability, and evaluation pipelines
• Tooling integrations across CRMs, ticketing systems, data warehouses, internal APIs, and document stores

Why this role is different

• You’ll build at the edge of what’s possible with agentic AI— in real production environments , not demos.
• You’ll influence product direction by feeding real deployment learnings back into Lyzr.
• You’ll own end-to-end delivery: requirements → architecture → build → deploy → iterate.

Compensation

• Base salary: Up to $150,000 USD/year
• Additional compensation may include equity and benefits (details shared during the interview process).
• (Base range depends on experience, scope, and customer deployment expectations.)