Software Engineer, AI & Data

CrowdHealth is hiring a mid-level software engineer to build and own backend services, data pipelines, and LLM-powered features in production. The role works across AI, data engineering, and backend development, partnering with product, operations, and data teams to deliver improvements.

Responsibilities:

• Build and maintain TypeScript and Python backend services and REST APIs
• Design batch data pipelines with orchestration, tests and data quality checks
• Build LLM features end to end: prompts, retrieval, structured outputs and evaluation, tracking accuracy, latency and cost
• Turn scanned documents into structured data using OCR and LLMs
• Containerize, deploy and monitor services in the cloud, and share on-call [rotation details]
• Write tests, review code and document what you build
• Work with product, operations and data partners to scope work and ship in small, frequent increments

Requirements:

• 3 to 5 years of professional software engineering experience, or equivalent
• Production experience with the Required technologies in the tech stack above
• Has shipped at least one LLM-powered feature to real users, with a way to measure its quality
• Clear written communication, including explaining technical tradeoffs to non engineers
• Languages: TypeScript and Python (strong), SQL (solid)
• Backend and APIs: REST APIs in Node.js or Python (e.g., Express, NestJS, FastAPI or Django); background jobs and queues
• Cloud and infrastructure: AWS, GCP or Azure; Docker; CI/CD
• Data engineering: Batch pipelines with Airflow, Dagster or dbt; data quality testing
• Applied AI and LLMs: LLM APIs (OpenAI, Anthropic or similar); prompt design; measuring output quality
• Observability: Logging, metrics and alerting for production services
• Experience with an AI-driven SDLC: using coding agents and assistants (e.g., Claude Code, Cursor or GitHub Copilot) across planning, coding, testing and code review
• Experience handling sensitive data (PII, financial or health records) under compliance programs like SOC 2
• Small-team experience owning systems end to end, from design through on-call
• Backend and APIs: Event-driven systems (Kafka, SQS or Pub/Sub)
• Cloud and infrastructure: Kubernetes, Terraform, Pulumi
• Data engineering: Snowflake, BigQuery or Redshift; Metabase
• Applied AI and LLMs: RAG and vector search (pgvector, Pinecone); OCR and document AI (Textract, Google Document AI, Azure Document Intelligence)
• Observability: Sentry, PostHog or Datadog