JOB-1721
An applied AI lab is building autonomous agents that work inside heavily customized enterprise resource planning systems, turning legacy back-office software into something that can execute finance and operations work with minimal human support. Its customers are large global companies, and the team has a background at leading AI labs and enterprise software vendors. It is hiring a founding engineer to take charge of a brand-new product line and its supporting infrastructure.
Remit
- Architect the agent layer, adding checks, reflection loops and validation steps ahead of anything reaching users or connected systems
- Orchestrate multi-step agent pipelines and retrieval-augmented generation (RAG)
- Set up low-latency inference paths with observability down to the token level
- Get prompt chains, document retrieval and targeted fine-tuning running reliably in production when they help
- Partner with customers on projects where your work has outsized effect, and influence team culture, engineering standards and roadmap
Good-fit signals
- One to six years as a software engineer, ML engineer or researcher; two to four is the sweet spot
- Production code shipped in a team codebase, beyond research or modeling code alone
- Solid ML fundamentals through research, internships or full-time work
- Experience with LLM platforms, agent harness engineering and model or agent evaluation
- Ownership mindset, crisp communication, and comfort deciding with incomplete information
- Eagerness to pick up unfamiliar technology stacks and use AI tools day to day; enterprise or B2B AI experience is a plus
On-site in the San Francisco Bay Area. Visa transfers and new visa sponsorship are both available.
