AI Engineer 3

As an AI Engineer, you will develop the automation systems that power Hexalog's AI platform. Building on the foundational automation workflows created by the team, you will architect more complex multi-step systems that coordinate LLM reasoning, browser automation, document intelligence, and workflow orchestration. You will be responsible not only for building automation agents, but for designing the systems that allow them to operate reliably in production, including orchestration logic, evaluation frameworks, and observability pipelines. You will also collaborate with other AI and backend engineers and help them to build automation tools into scalable platforms.

Responsibilities:

• Automation platforms that execute complex operational workflows across multiple business verticals.
• Multi-agent automation systems that interact with external software through browser automation, APIs, and document intelligence pipelines.
• Workflow orchestration frameworks that coordinate multi-step automation pipelines across agents, services, and human-in-the-loop escalation paths.
• Evaluation and observability systems that measure LLM reasoning quality, extraction accuracy, and end-to-end workflow reliability.
• Backend services and data pipelines that maintain system state and ensure reliable execution of automation workflows in production.
• Foundations for scalable internal automation systems that will evolve into reusable platforms across the company.

Requirements:

• 5+ years of experience building and owning distributed backend systems or AI-driven platforms in production, with a track record of taking systems from prototype to reliable production.
• Deep proficiency in Python and TypeScript; strong experience designing, shipping, and maintaining. production-grade backend services and APIs (FastAPI or similar).
• Proven experience in architecting multi-agent systems, designing agent coordination patterns, managing shared state, handling failure modes, and ensuring reliable execution across long-running workflows.
• Experience designing or operating workflow orchestration systems (Temporal, Airflow, or similar) at a level where you can make principled decisions about retry logic, idempotency, and pipeline reliability.
• Strong familiarity with evaluation and observability for AI systems, including designing evaluation frameworks, tracking LLM reasoning quality, extraction accuracy, and end-to-end workflow reliability in production.
• Hands-on experience with cloud-native infrastructure (AWS Lambda, Fargate, ECS, or equivalent) and Infrastructure as Code; comfortable owning the deployment architecture for AI workloads.
• Experience with LLM APIs, tool use, and agentic frameworks; able to reason about prompt design, context management, and model behaviour at a systems level.