AI Engineer

About Nomain

Nomain is a deep-tech startup accelerating legacy code modernization. Our AI-powered platform extracts decades of hidden insights from legacy mainframe codebases and presents them through an intuitive interface for developers, business analysts, and architects alike.

We believe AI will accelerate experts, not replace them. Great software systems are built by people who have the right context and the right tools. Our product delivers that specific context while leaving the ultimate decisions in the hands of the experts.

Our Culture

At Nomain, we are customer-obsessed. We work closely with our users to understand their challenges and build solutions that truly make a difference. Our culture is built on ownership, transparency and continuous learning. We’re a tight-knit, ambitious team that values curiosity, autonomy, and building things that matter. Everyone has a voice, and we’re united by a shared mission to solve complex problems with practical and impactful solutions.

The Opportunity

We’re building the AI core of the Nomain Knowledge Platform: the systems that read, understand, and reason about codebases that have been growing for decades. The work sits at the intersection of language models, retrieval, and agentic systems, and it is the part of the product where the next year of progress will define how far we can go.

We’re looking for a Senior AI Engineer to push the limits of what our platform can do. You’ll design and build coding agents and software factories, deepen our code analysis with state of the art retrieval and reasoning, and explore self-improving agents. The bet we’re making is that small teams using AI-native development, with deliberate context and well-designed agents, can outbuild much larger ones. We use Claude Code daily and treat AI tooling as part of the engineering stack, not a side experiment.

You’ll also collaborate with the team on improving our internal development practices, raising the bar on how we use AI in our own workflows, and refining the product alongside customers.

This role reports directly to the CTO and requires close collaboration with the founding team and customers.

What We’re Looking For

This role requires real depth in LLMs and agentic systems, combined with the engineering discipline to ship them in production. You should understand how language models work under the hood, have built agents that survive contact with real users, and be fluent in AI-native development. You move quickly across the stack of model, context, tools, and evaluation, while keeping sight of what actually helps customers.

Technical requirements

• 5+ years of software engineering experience, with at least 2 years working with LLMs and agentic systems in production. Or proof that experience is not everything.
• Deep understanding of how LLMs work: tokenization, attention, embeddings, context windows, inference economics, and fine-tuning trade-offs.
• Hands-on experience designing and shipping agents end-to-end: tool design, planning, memory, evaluation harnesses, and failure-mode handling.
• Strong intuition for context engineering and retrieval, and for when to lean on prompting, retrieval, fine-tuning, or model choice.
• Fluent in AI-native development with Claude Code or similar agentic coding tools, used as a daily part of how you build.
• Comfortable working in Python, TypeScript and/or C#. But more importantly, know what to do, with AI by your side the language is less important.
• Experience deploying and iterating on AI systems in production, with evals you actually trust.

Highly valued

• Search and retrieval: hybrid retrieval, reranking, grounding LLMs in large corpora, building or operating vector databases.
• Deep code analysis: program representations, language compilers, grammars, ASTs, or knowledge graphs.
• Experience building self-improving systems: agents that learn from their own runs, reinforcement loops, automated evals.
• Multi-agent orchestration and software factories: pipelines where multiple agents and humans collaborate on real software.
• Research background or active engagement with the AI literature, balanced with a pragmatic shipping mindset.
• Experience with OpenSearch, PostgreSQL and pgvector, or comparable vector and hybrid retrieval stacks.

Beyond technical skills, you bring the mindset that aligns with our culture

• Customer-obsessed: You build AI capabilities to help users accomplish things. You think about impact on user experience and work backwards from their needs.
• Solution focused: You see opportunities for improvement. When you spot problems, you suggest better approaches rather than just flagging issues.
• Transparent communication: You share your reasoning, document decisions, and aren’t afraid to say when requirements don’t match reality.
• Continuous learner: The AI field moves fast and you keep up. You learn from model releases, papers, your own evals, and share those learnings with the team.
• Curious and autonomous: You ask why agents fail the way they do and take initiative to make them better. You gather context and move forward.
• Creative: You combine ideas in unexpected ways. You design agents and workflows that look obvious only after you’ve built them.
• Practical problem solver: You focus on systems that work and help users, not impressive demos that fall over on the second prompt.
• Collaborative contributor: You work with users, engineering, and product teams. You make better decisions by understanding different perspectives.

Why This Role

• Frontier problems: Coding agents, software factories, self-improving systems, deep code analysis. These are open problems and you’ll work on them with full context.
• Full ownership: You own the AI core, end to end. Models, context, tools, agents, evals. You make the decisions and see them ship.
• Direct impact: Your work is the product. There’s no separation between research and what customers see.
• AI-native team: We use Claude Code and modern AI tooling daily. You’ll be in a team that takes leverage seriously and expects you to push it further.
• Growth path: As we scale, you may help build and lead AI engineering at Nomain. Work directly with the CTO on architecture and product direction.

Technical Environment

• AI stack: frontier LLMs, embeddings, hybrid retrieval, vector databases (pgvector), agent frameworks, evals, prompt and context tooling.
• Programming languages and frameworks: Python, .NET / C#, TypeScript.
• Cloud and infrastructure: Microsoft Azure, AWS, PostgreSQL, Containers, Kubernetes, APIs and API management.
• Development practices: TDD, Merge/Pull Request, CI/CD, Infrastructure-as-Code, GitOps, AI-assisted coding (Claude Code), LLM Evals.