Forward Deployed Engineer (FDE)

Role Overview As Forward Deployed Engineer at Impetus, you will own the intelligence layer powering enterprise workflows. You will design, build, and optimize agent-driven systems that operate at scale—handling transactions, making decisions, resolving exceptions, and continuously learning from real-world usage. This role is for engineers who have gone beyond experimentation and have built AI systems that run reliably in production. You understand how to turn the potential of large language models into systems that enterprises can trust. What You’ll Do Design and build robust agent architectures for complex enterprise workflows, including multi-step reasoning, tool usage, and exception handling Develop and maintain evaluation systems to ensure accuracy, reliability, safety, and consistency of AI outputs Build prompt strategies, retrieval pipelines, and context engineering frameworks for predictable agent performance Create feedback loops that enable systems to learn from human input and improve over time Optimize performance across latency, throughput, and cost for production environments Establish best practices for observability, governance, and reliability of AI systems Stay current with advancements in models, frameworks, and techniques—and bring relevant innovations into production What We’re Looking For 7+ years of software engineering experience, with at least 2 years focused on building AI or LLM-based systems in production Hands-on experience developing agentic workflows involving tool calling, retrieval, and multi-step reasoning Strong understanding of prompt engineering, context design, and reliability challenges in LLM systems Experience building evaluation frameworks for AI performance and quality Strong programming skills in Python and solid backend engineering fundamentals Proven track record of shipping AI features and handling real-world challenges such as edge cases, hallucinations, and cost management High ownership mindset with the ability to independently drive systems from concept to production Helpful Experience Experience with agent frameworks (e.g., LangGraph, CrewAI, or custom orchestration systems) Building retrieval systems using vector databases (e.g., Qdrant, pgvector, Pinecone), including hybrid search and reranking Familiarity with tool-calling protocols, model context protocols, and API integrations Experience with model fine-tuning or adaptation techniques (e.g., LoRA) Exposure to evaluation pipelines and continuous monitoring systems Prior experience deploying AI solutions in enterprise environments (including compliance and governance considerations) Background in AI-native startups, applied ML teams, or similar environments Why Impetus Be part of a high-caliber, early-stage team shaping how AI is deployed in real enterprise environments Work on real-world problems where your systems directly drive business impact Operate at the intersection of engineering, product, and strategy High ownership, fast execution, and meaningful influence on the company’s direction Show more Show less