Backend Engineer, AI Engine; Python

Department: Data Science and AI

Employment Type: Full Time

Location: Remote - Istanbul, Türkiye

Reporting To: Davide Fornelli

Surgo Health is building the Behavioral Intelligence layer for healthcare - revealing why people behave as they do, predicting what they will do next, and helping organizations make the decisions that change outcomes.

A Public Benefit Corporation grounded in two decades of global health experience, Surgo generates foundational behavioral data that reveals the unseen why and builds the AI products that power decision-grade insights at population scale - equipping healthcare organizations, funders, and public agencies to act on what actually shapes health outcomes: people's beliefs, motivations, and barriers.

Backed by 40+ peer-reviewed publications and trusted by partners including the Gates Foundation, J&J, Takeda, Pivotal Ventures, and the Clinton Health Access Initiative, Surgo is building a future where healthcare organizations don't just understand human behavior, but anticipate it.

We are looking for a Backend Engineer to join the R&D Engineering team building Derin, our AI-powered interview platform. At Derin's core is an interview engine: an LLM agent that conducts conversational interviews with participants over web, Whats App/SMS and voice, follows a researcher-designed survey with branching, quotas and screening, probes for depth, and keeps every conversation safe.

You will own major parts of this engine: the survey definition language and its execution, the agent's conversational behavior, the APIs and data that the rest of the platform depends on, and the cloud infrastructure it runs on. This is a hands-on role with real ownership, in a small team where your work reaches thousands of participants quickly.

• Work AI-first: use AI agents (Claude Code, Cursor and others) throughout your day to design, build, debug and optimize, and help the team keep raising the bar on AI-assisted engineering
• Build and scale the backend of our AI interview engine in Python (FastAPI, serverless on Azure), from API design to production
• Evolve how our surveys are defined and executed, with strong validation
• Shape how our LLM interviewer behaves, and evaluate changes rigorously before they ship
• Integrate messaging, voice and participant-recruitment channels
• Design data models and migrations on PostgreSQL, and expose engine data cleanly to our dashboards, exports and data platform
• Own reliability and observability: automated tests (including deterministic tests of LLM flows), monitoring, tracing and alerting
• Work with infrastructure as code (Terraform), CI/CD and secrets management on Azure
• Build security, privacy and compliance into how we handle sensitive participant data, in a SOC 2 and HIPAA-compliant environment
• Write short design documents for significant changes, and review your teammates' work
• Collaborate closely with product, research scientists, frontend and data engineers

• You have at least 3 years of backend development experience in Python
• You have proven experience designing and running APIs in production (FastAPI or similar)
• You are strong with relational databases, particularly PostgreSQL, and with ORMs and migrations (SQL Alchemy/SQLModel, Alembic)
• You have built LLM-driven applications in production: prompt design, structured outputs, tool use or agents, and evaluating non-deterministic behavior
• You have hands-on experience with Microsoft Azure (required), including serverless or event-driven architectures (queues, functions)
• You understand and value software engineering best practices: testing (pytest), code review and CI/CD
• Experience with Terraform or with conversational and voice products is…