Skip to main content

LangChain

Fullstack Software Engineer, Applied AI

San Francisco, CA$165k–$190kfulltimemidAdded 2 days ago

About this role

LangChain is seeking a Fullstack Applied AI Engineer to design and deploy production-grade AI agents and workflows across the company's operations and open-source ecosystem. You'll work at the frontier of agent engineering, collaborating with cross-functional teams to build autonomous systems that accelerate LangChain's mission while contributing to widely-used frameworks like LangChain and LangGraph.

What you'll do

  • Design and deploy end-to-end AI workflows and agents solving real problems across multiple business domains
  • Develop agent architectures, evaluation pipelines, and performance frameworks to ensure reliability
  • Translate emerging AI research into practical, production-ready solutions
  • Collaborate with cross-functional teams (Marketing, GTM, Recruiting, Product) to identify automation opportunities
  • Contribute to LangChain and LangGraph ecosystem including open source components and documentation
  • Communicate technical decisions and trade-offs to technical and non-technical stakeholders

What they're looking for

  • Python or TypeScript (ideally both)
  • LLM systems and production AI applications
  • Evaluation and monitoring systems for agents
  • Prompting, retrieval, orchestration, and model selection
  • Agent architecture and workflow design
  • Clear technical communication
  • Fast-paced startup problem-solving
  • Curiosity about emerging AI tools and frameworks

Benefits

  • Work at the frontier of agent engineering and AI research
  • Contribute to widely-adopted open-source projects (100M+ monthly downloads)
  • Meaningful impact on company direction and product development
  • 5-day in-office work in San Francisco or New York (hybrid eligibility based on location)
  • Opportunity to ship production systems used by Fortune 500 companies
Apply on the employer's site

Opens the official application on the employer’s site. No login required.