Role Overview
We are seeking a Senior AI/LangGraph Engineer to design, build, and productionize intelligent agent systems powered by large language models. This role will lead the development of reliable, observable, and secure multi-agent workflows using Python, LangGraph, FastMCP, and modern prompt-engineering practices.
The ideal candidate combines strong software engineering fundamentals with practical experience building LLM applications. You will work across architecture, implementation, evaluation, and operations to turn complex business and technical requirements into scalable AI solutions.
Key Responsibilities
• Design and implement stateful, graph-based agent workflows using LangGraph and Python.
• Architect multi-agent systems with clear delegation, coordination, memory, tool use, and human-in-the-loop patterns.
• Build and integrate Model Context Protocol (MCP) servers and tools using FastMCP.
• Develop robust prompts, system instructions, structured-output strategies, and reusable agent behaviors.
• Create agent routing, planning, reflection, retrieval, and task-execution patterns that are reliable in production.
• Integrate LLMs with enterprise APIs, databases, vector stores, observability platforms, and business applications.
• Establish evaluation frameworks for agent quality, task completion, groundedness, latency, safety, and cost.
• Implement testing strategies for prompts, tools, workflows, failure modes, regressions, and adversarial inputs.
• Build safeguards for permissions, data privacy, prompt injection, unsafe tool use, and unauthorized actions.
• Improve system performance through model selection, caching, concurrency, context management, and token optimization.
• Partner with product, data, platform, security, and domain teams to define requirements and deliver usable solutions.
• Mentor engineers and contribute to coding standards, architecture patterns, technical documentation, and engineering practices.
Required Qualifications
• 5+ years of professional software engineering experience, including substantial production experience with Python.
• Demonstrated experience building LLM-powered applications, AI agents, or workflow automation systems.
• Hands-on experience with LangGraph or comparable graph/state-machine orchestration frameworks.
• Strong understanding of multi-agent architecture, including agent boundaries, coordination, tool calling, state, memory, and failure recovery.
• Experience with prompt engineering, including instruction design, few-shot examples, structured outputs, tool-use prompts, and prompt evaluation.
• Experience building MCP servers or tools, preferably with FastMCP.
• Proficiency designing clean APIs, reusable components, asynchronous workflows, and production-grade services.
• Experience integrating REST or event-driven services, databases, authentication, and third-party APIs.
• Understanding of LLM limitations, including hallucination, context constraints, nondeterminism, latency, and cost management.
• Strong debugging, testing, code-review, and technical communication skills.
Preferred Qualifications
• Experience with LangChain, LangSmith, OpenAI-compatible APIs, Anthropic APIs, or other leading LLM platforms.
• Experience with retrieval-augmented generation, embeddings, vector databases, document processing, or knowledge graphs.
• Experience deploying AI services on AWS, Azure, or Google Cloud.
• Familiarity with Docker, Kubernetes, CI/CD, infrastructure as code, and cloud-native observability.
• Experience with streaming responses, durable execution, workflow persistence, queues, and distributed systems.
• Knowledge of model fine-tuning, synthetic data generation, or automated evaluation methods.
• Experience implementing role-based access control, secrets management, audit logging, and responsible-AI controls.
• Contributions to open-source projects, technical communities, or internal AI platforms.
What Success Looks Like
• Agent workflows are reliable, testable, observable, and maintainable in production.
• Multi-agent systems complete complex tasks with appropriate tool use and clear escalation paths.
• Prompts and agent behaviors are versioned, evaluated, and improved through measurable feedback loops.
• MCP tools expose secure, well-documented capabilities with predictable schemas and failure handling.
• AI solutions meet agreed targets for quality, latency, availability, safety, and cost.
• Engineering teams can reuse proven patterns, components, evaluation methods, and operational playbooks.
Core Competencies
• Systems thinking and pragmatic architecture
• Strong Python engineering
• AI-agent and workflow design
• Prompt and context engineering
• Production reliability and observability
• Security-minded development
• Clear written and verbal communication
• Mentorship and technical leadership
Never miss an opportunity! Create an alert based on the job you applied for.
