Forward Deployed AI Engineer

This is a fully remote position, open to applicants in United Kingdom.

📋 Description

• Engage within the client's environment from the discovery phase through to architecture, development, production, and handover.

• Design and develop production-grade agentic AI systems tailored for enterprise clients.

• Execute multi-agent orchestration, including tool and function invocation, retrieval, planning, routing, human-in-the-loop checkpoints, state management, guardrails, and failure recovery mechanisms.

• Create ingestion and enrichment pipelines for both structured and unstructured data, encompassing documents, PDFs, and object storage.

• Set up evaluation datasets, automate grading processes, and monitor groundedness, faithfulness, and retrieval quality metrics.

• Enhance model selection, routing strategies, prompt and context budgets, caching solutions, cost efficiency, and latency optimization.

• Provide observability, tracing, CI/CD practices, infrastructure as code, monitoring, security, and compliance controls.

• Collaborate closely with client architects, DevOps teams, QA specialists, and engineers within the client's codebase and cloud environment.

• Facilitate knowledge transfer through pairing, documentation, monitoring handover, and client enablement initiatives.

• Participate in scoping, estimation, pre-sales activities, reusable asset development, and interviewing as relevant to the position level.

⛳️ Requirements

• Proficient in Python, including async programming, typing, testing, and packaging.

• Familiarity with TypeScript / Node is advantageous.

• Practical production experience with advanced models, prompt and context engineering, structured outputs, tool and function invocation, streaming, and token/context-window management.

• Proven experience in developing and deploying multi-agent or agentic workflows, planner/router patterns, supervisor and sub-agent architectures, ReAct-style loops, state and checkpointing systems, retries, interrupts, and human-in-the-loop review processes.

• Experience with the Anthropic Agent SDK, LangGraph, LlamaIndex, or comparable orchestration frameworks.

• Understanding of chunking, embeddings, hybrid and semantic search, re-ranking, citation and provenance management, and translating natural language into queries.

• Experience with NoSQL databases like MongoDB or similar, along with SQL; adept at designing schemas for agent retrieval.

• Background in constructing ingestion and enrichment pipelines from complex structured and unstructured data sources.

• Experience in creating evaluation harnesses and datasets, LLM-as-judge systems, regression testing, and use-case specific metrics.

• Hands-on production delivery experience on AWS, including Bedrock, Azure, or GCP; knowledge of containers, serverless architecture, networking basics, secrets management, and IAM.

• Proficient in Git, code review processes, testing, CI/CD, infrastructure as code, and observability practices.

• For Mid FDE: 3+ years in software engineering with a minimum of 1 year of hands-on experience in LLM/agentic work, including at least one production system.

• For Senior FDE: 5+ years in engineering with at least 2 years in AI or agentic AI, including complete ownership of the architecture for a production agent system.

• For Lead FDE: 7+ years of experience, including leadership in delivery teams, technical leadership in engagements, pre-sales, estimation, and mentoring.

• Strong client-facing skills, effective technical and business communication, ability to translate business challenges into technical solutions, collaboration skills, tolerance for ambiguity, focus on production, teaching capabilities, ownership mentality, commercial awareness, resilience, and proficient written communication in English.

• Willingness to travel occasionally.

• Anthropic certification is a plus, but not mandatory.

🏝️ Benefits

• Support for obtaining vendor certifications.

• Access to partner training programs.

• Company-wide AI tools integrated into everyday work practices.