Applied AI Engineer – Agentic Workflows

Cohere Cohere · AI Frontier · San Francisco, CA · Modeling

Cohere is seeking an Applied AI Engineer to build production-grade AI agents for enterprise customers. This role involves designing, building, and deploying agentic workflows powered by LLMs, integrating them with tools, APIs, and data sources. The engineer will focus on reliability, observability, safety, and audibility, working closely with customers and shaping how agentic systems are built and deployed.

What you'd actually do

  1. Work closely with enterprise customers to translate high-value, ambiguous business problems into well-framed agentic problems with clear success criteria and evaluation methodologies.
  2. Provide technical leadership across the full development and evaluation lifecycle, including post-deployment iteration, for agentic workflows.
  3. Contribute to shared frameworks and patterns that enable consistent delivery across customers.
  4. Lead the design, build, and delivery of LLM-powered agents that reason, plan, and act across tools and data sources with enterprise-grade reliability and performance.
  5. Balance rapid iteration with enterprise requirements, evolving prototypes into stable, reusable solutions.

Skills

Required

  • Python/TypeScript
  • building, shipping, and maintaining production-grade software
  • building agents that plan and execute multi-step tasks
  • interacting with external APIs/tools
  • Frontier Models (GPT, Claude, Gemini)
  • RAG
  • vector databases (Pinecone, Weaviate, etc.)
  • orchestration frameworks (LangGraph, CrewAI, or custom state machines)
  • building robust evaluation frameworks to measure agent accuracy, safety, and latency
  • leading technical discussions with enterprise customers
  • mentoring distributed teams
  • setting the architectural standards for AI/Agentic systems

Nice to have

  • Strong written and verbal communication skills.
  • Ability and interest to travel up to 25%, flexible.

What the JD emphasized

  • production-grade AI agents
  • agentic workflows
  • reliable, observable, safe, and auditable
  • robust evaluation frameworks

Other signals

  • production-grade AI agents
  • agentic workflows
  • LLM integration
  • enterprise customers