SoftwareEngineer,AppliedAI

ABOUT US AND THE ROLE At Sobek AI, we’re building secure AI infrastructure for agentic workflows in life-sciences innovation networks and intergovernmental emergency response. Backed by $10M+ in grants and funding, we work with global, high impact partners on distributed workflows where reliability, security, and trust matter from day one. Our systems are already deployed in mission-critical customer environments. We’re hiring a Software Engineer to help build the production AI systems behind Sobek’s core offerings, sitting where agentic workflows meet enterprise data and trust boundaries. This is a foundational role on the engineering team, so we’re looking for someone who has shipped AI systems used by real users and has the software judgment to harden them for sensitive data and scale performance. This means experience with defining clear access boundaries, measurable quality, failure handling, and debuggable interfaces. While this is not a research role, it does require practical ML and LLM fundamentals. You should understand enough about how models are trained, evaluated, served, and deployed to make sound engineering decisions when building with them. WHAT YOU’LL DO BUILD PRODUCTION AI WORKFLOWS - Build agentic workflows over enterprise and government data, with clear rules for what a model can see, what tools it can call, and when a human needs to review or approve an action. - Design context and grounding systems that give models the right information at the right time without violating permissions or performance constraints. - Work across backend services, APIs, async workers, data pipelines, internal tools, and product facing surfaces. ENGINEER RELIABLE LLM SYSTEMS - Build evals and feedback loops for model behavior and workflow outcomes. - Own tracing and runtime visibility across models, context, tool calls, generated outputs. - Debug failures from evidence: context, traces, tool responses, user review, production logs. - Improve quality without ignoring