The role is hands-on backend work in a small team of four or five, building and shipping features that turn insurance risk information into structured data. A strong candidate is comfortable with modern typed Python, building and documenting REST APIs, working with Postgres, and running serverless code on AWS. Experience with Docker, message queues and email or notification services helps, and so does a habit of owning a feature from prototype through to support.
How it fits the AI era
Generative AI sits at the core of the product, so much of the backend work involves wiring model outputs, document reading and validation into dependable services. AI can speed up drafting code and tests, but judgment about data quality, failure handling and what an insurer can safely rely on stays with the engineer. Strong typing and clear API contracts matter more when model outputs are unpredictable.
How it builds AI experience
Working here means building production systems around large language models, including document extraction with OCR tools and Gemini on Google Cloud. Someone without this background could build it by making a small project that reads documents, extracts fields into a validated schema and tracks where the model gets it wrong. Practising prompt design and checking results against known answers would be a good start.
Worth showing
- Typed Python with mypy
- Serverless AWS Lambda projects
- REST API design with OpenAPI
- Postgres schema and queries
- LLM or OCR pipeline work
This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.
About Cytora
Cytora builds a digital underwriting platform that digitizes commercial insurance submissions, enriches exposure data, and automates triage decisions.
#J-18808-Ljbffr
