Applied AI Engineer

About the role

Work spans the whole lifecycle of an AI feature: experiment, evaluate, deploy and keep watching it. The right person thinks about latency, cost and quality in production, not just about models.

At a glance

What you will do

• Ship tool-calling LLM agents with structured output, parallel API calls and streaming
• Build evaluation uses and scoring rubrics for safety, usefulness and personalization
• Take a capability from prototype through evals, staging, production and trace monitoring
• Cut latency and cost with prompt caching, token budgets, retries and observability
• Write the API tools agents call, with Pydantic validation and database search

What they look for

• Strong Python, with services deployed on large production systems
• Experience with LangChain, LangGraph or a similar agent framework
• Hands-on LLM work in production: prompting, function calling, structured output, evaluation
• Comfort with async Python, HTTP APIs and streaming such as SSE or webhooks
• Data validation and schema design with Pydantic or JSON Schema
• Debugging across layers, from a failed tool call to a Terraform misconfiguration

Nice to have

• AWS services such as Bedrock, ECR, CloudFront, S3 or Cognito
• Tracing and eval tooling such as Langfuse, OpenTelemetry, Datadog or Terraform

Benefits

• Base salary plus equity