AI Engineer | EMEA/LATAM

About the role

You will own the design and delivery of AI features that directly influence how government buyers discover and engage with public sector opportunities. You will architect systems that ingest fragmented public signals and transform them into structured, actionable intelligence for our platform. This role requires you to evaluate and fine-tune language models so they perform reliably on complex policy and procurement data. You will collaborate with product and engineering teams to turn ambiguous requirements into production-grade AI capabilities. You will implement rigorous evaluation frameworks to measure model quality, latency, and cost in real-world scenarios. You will contribute to the foundational AI infrastructure that powers document analysis, semantic search, and conversational interfaces. You will play a key role in shaping how Starbridge leverages generative AI to create durable competitive advantages in the public sector market.

Key facts

What you'll do

• Architect and implement AI-powered features such as AI proposal writing and intelligent search experiences that serve public sector customers.
• Evaluate and continuously monitor the performance of models from OpenAI, Anthropic, Gemini, and Parallel.ai through structured testing and experimentation.
• Design and run experiments that quantify the impact of new AI capabilities on user engagement and pipeline generation.
• Collaborate closely with product managers to translate strategic requirements into technical approaches that maximize the value of LLMs for our platform.
• Implement robust testing strategies and CI/CD pipelines that enable rapid iteration while maintaining high reliability for production AI systems.
• Stay current with advances in AI and machine learning research and proactively identify opportunities to upgrade our generative AI stack.
• Partner with data scientists and infrastructure engineers to improve our model serving stack, latency, and cost efficiency.
• Document system behaviors, experiment results, and model performance metrics to support data-driven decision making.
• Contribute to the design of deep document analysis workflows that extract structured insights from RFPs, council records, and budget documents.
• Build and maintain integrations with external model providers and internal data pipelines to ensure scalable and secure AI workflows.

Requirements

• Hold a Bachelor's degree in Computer Science, Engineering, Mathematics, or a related field, or demonstrate equivalent experience through professional work history.
• Have professional experience in a startup or smaller company where your engineering contributions had direct and measurable impact on product outcomes.
• Bring a minimum of 5 years of professional experience in software engineering and AI/ML development across production software and systems design.
• Demonstrate proficiency with Python as a primary language for building and deploying machine learning systems in production environments.
• Show deep understanding of machine learning algorithms and model development techniques across supervised, unsupervised, and reinforcement learning paradigms.
• Have hands-on experience with ML lifecycle tools such as MLflow, DVC, and weights & biases to track experiments and model versions.
• Prove experience with cloud deployment of ML systems on major infrastructure platforms with strong operational reliability.
• Have professional experience working with large language models and building applications on top of complex, large-scale models.
• Show strong software engineering fundamentals with a track record of building scalable and distributed machine learning systems that serve real users.
• Demonstrate product thinking by taking ambiguous requirements and strategizing how language models can be applied to generate meaningful public sector insights.
• Communicate complex technical and product concepts clearly to both technical and non-technical stakeholders during reviews and discussions.

Nice to have

• Experience building scalable applications with LLMs using frameworks such as LangGraph, LiteLLM, Agent Client Protocol, and Koog.
• Deep knowledge of RAG implementation patterns and techniques to improve retrieval quality, relevance, and system efficiency.
• Proficiency with Kotlin for building backend services that integrate with AI workflows.

Practical notes

All interviews are conducted via Google Meet. The interview process is fast-paced, with prompt responses valued over delayed scheduling. Please keep us posted on your timeline so we can move quickly and speed things up where needed. This role is full-time and based in Europe or LATAM. Compensation details are not specified in this posting.