Forward Deployed AI Engineer

About the role

The Forward Deployed AI Engineer works directly with clients and senior engineers to build AI-powered systems inside real enterprise environments. It suits a strong junior engineer who wants to work on practical AI systems, enterprise data, agents, retrieval, workflow automation and context graphs. The work is hands-on and client-facing.

Key responsibilities

• Understand operational workflows and turn them into technical artifacts.
• Connect to real enterprise systems and data sources.
• Build context-aware AI agents, retrieval systems and automation prototypes.
• Ingest, clean, structure and connect data from enterprise systems into context graphs, including databases, PDFs, spreadsheets, CRM data, Slack or Teams exports and meeting transcripts.
• Develop AI and agentic workflows, including LLMs, retrieval systems, structured extraction, tool use, LangGraph-style workflows and agent harnesses.
• Integrate with client infrastructure such as databases, APIs, cloud storage, document repositories, analytics tools and ticketing systems.
• Rapidly build proof-of-concepts showing how AI can improve operations, such as workflow discovery, incident management, document extraction or customer journey analysis.
• Translate ambiguous business problems into technical designs, data models, prompts, pipelines and deployed applications.
• Help build reliable, maintainable systems that can run in production, including client cloud environments.

Required profile

• A strong junior software engineer interested in applied AI and real systems.
• Comfortable with incomplete, messy or poorly documented data and with moving quickly from vague requirements to a working prototype.
• Clear communication with engineers and non-technical stakeholders.
• Curious about how AI agents interact with real tools, data and workflows.

Required skills

• Strong Python fundamentals and basic backend development experience.
• Familiarity with APIs, databases and cloud services.
• Interest in LLMs, agents, retrieval, structured outputs and tool calling.
• Experience with data pipelines and messy real-world datasets.
• Plus: FastAPI, LangChain, LangGraph, vector databases, document processing, knowledge graphs, AWS, and enterprise data, analytics or workflow automation.