Technical Lead, AI

Reporting to: BE Tech Lead

Closest Collaboration: Hume Engineering Team, Product Management, Chief Scientist

Job Location: Open to candidates from anywhere in the EU and UK

Flexibility: Option to work fully remotely

Keywords: LLM, generative AI, LangChain, LangGraph, CI/CD pipelines, Java, Python, OpenAI, HuggingFace, vector stores

The mission

We are looking for a hands-on AI Engineer to help build and integrate Generative AI and Large Language Model (LLM) capabilities into Hume, our connected data analytics platform used by intelligence analysts in mission critical environments. You will work embedded in an existing engineering team, bringing deep practical experience with LLMs, agentic systems, and production-grade software development.

Your primary focus will be Hume Maestro, our core AI component — the layer that brings intelligence to a real-world platform already trusted by government agencies to make sense of complex, connected data. Beyond building, you will be a key enabler for the wider engineering group, shaping how AI becomes a first-class capability within Hume: sharing knowledge, defining best practices, and guiding colleagues on how to responsibly integrate LLM-based and Generative AI solutions into a product that genuinely matters.

Key responsibilities

• Agentic System Design: Architect and implement agentic AI workflows, including patterns such as tool use, reflection, multi-step reasoning, and prompt chaining, leveraging frameworks like LangChain and LangGraph.
• Hands-on AI Development: Design, build, and iterate on LLM-powered features within the Hume platform, with a strong focus on quality, performance, and maintainability.
• Engineering Integration: Work alongside existing product engineers to integrate Generative AI capabilities into current product components, ensuring seamless fit with established codebases and deployment pipelines.
• Developer Enablement: Act as an internal expert and mentor — helping other engineers understand LLM integration patterns, prompt engineering, and evaluation techniques, raising the team’s overall GenAI capability.
• Deployment & Operations: Support the deployment lifecycle of AI components, collaborating on CI/CD processes, model serving, monitoring, and reliability in production environments.
• Technical Quality: Drive engineering best practices specific to AI: prompt versioning, evaluation frameworks, regression testing for model outputs, and robust observability.

You will be the first dedicated AI Engineer at GraphAware, joining an established engineering team and a product with real customers. Hume Maestro is early enough that you will shape its architecture, framework choices, evaluation strategy, and the engineering practices the wider team adopts. 

• Unique opportunity: Work on technology that genuinely matters — helping democratic government agencies and law enforcement analysts solve cases faster, make better decisions, and protect the communities they serve
• Driven and motivating environment: Striving for excellence is who we are but we’ve always got each other’s back. Work in an environment that motivates and supports you
• Self-realisation and autonomy: Thrive in an environment where autonomy is balanced with accountability. If something isn’t working, we take action
• Competitive compensation: Enjoy a competitive compensation package and regular reviews
• Equity participation: Share in the company's success through our share scheme program

What you need to bring

Engineering foundation

• Experience: At least 5 years of hands-on software engineering experience in a professional engineering team, working on real, production-grade products.
• Programming: Strong development skills in Java or Python (either is equally valued). Demonstrated ability to write clean, tested, and maintainable code.
• Engineering mindset: Comfortable working within established engineering workflows — version control, code review, CI/CD pipelines, and deployment processes.

Generative AI & LLM expertise

• LLM experience: At least 2 years working with Large Language Models in a product or engineering context — not just experimentation, but shipping features.
• Agentic systems: Hands-on experience building agentic AI configurations (ideally using LangChain and/or LangGraph), with knowledge of patterns such as reflection, tool use, and multi-agent orchestration.
• Prompt engineering: Practical knowledge of prompt design, optimisation, and evaluation, including systematic approaches to measuring and improving prompt quality.
• Frameworks: Proficiency with LangChain and/or LangGraph. Familiarity with broader GenAI tooling (e.g. OpenAI, HuggingFace, vector stores) is a strong plus.

Collaboration & knowledge sharing

• Team integration: Proven ability to work within cross-functional engineering teams, communicating effectively with both technical and non-technical stakeholders.
• Enablement: Experience supporting or informally mentoring other engineers in adopting new technologies or approaches.
• Product mindset: Ability to translate AI capabilities into practical product value, understanding trade-offs between innovation and delivery timelines.

• On-premises deployment: Experience delivering software in on-premises environments, including containerised deployments using Docker and related tooling.
• Graph data models: Familiarity with graph-based data models or connected data analytics.
• AI evaluation: Prior involvement in AI evaluation frameworks or benchmark design.
• RAG architectures: Experience with Retrieval-Augmented Generation (RAG) architectures and vector stores.

GraphAware’s core values

• Sense of Ownership
• Strive for Excellence
• Driven by Customer Success
• Belief in the Value of Graphs
• I Got Your Back!
• Radical Candor