Simplify Logo
Intent Lab

Intent Lab

Autonomous AI software engineering platform

Agent Engineer / Forward Deployed Engineer

Full-Time
No salary listed
Mid
Sunnyvale, CA, USA
In Person

Onsite role; candidates must be based in the Bay Area or willing to relocate.

About the job

Requirements
  • At least 3 years of software engineering experience with a strong foundation in backend or infrastructure work.
  • Hands-on experience with large language models and agent systems, including prompt engineering, tool use, evaluations, and multi-step orchestration.
  • Familiarity with distributed systems, containers, continuous integration and continuous delivery, and cloud platforms such as Amazon Web Services, Google Cloud Platform, or Microsoft Azure.
  • Proficiency in at least one modern programming language, such as Python, TypeScript, or Go.
  • Strong written and verbal communication skills, including comfort working with senior engineers and executives.
  • Ability to work independently in ambiguous environments, handle context switching, and sustain work in high-intensity conditions.
  • Based in the Bay Area or willing to relocate for an onsite role.
Responsibilities
  • Lead end-to-end technical deployments of IntentLab inside customer engineering organizations.
  • Integrate the system into customer codebases, continuous integration and continuous delivery pipelines, and cloud infrastructure.
  • Own the customer technical lifecycle from first integration through steady-state operation.
  • Build and tune custom agent behaviors, tool integrations, and evaluation harnesses for each customer's backend.
  • Debug agent performance across distributed systems and unfamiliar production code.
  • Validate integrations and maintain reliable system performance in customer environments.
  • Lead demos, pilots, and technical workshops with engineers, platform leads, and chief technology officers.
  • Build relationships with customer engineering teams and drive adoption from within those teams.
  • Translate customer pain and usage patterns into product and research priorities.
  • Ship fixes and features directly rather than only filing tickets.
  • Write documentation, reference integrations, and playbooks that turn one-off wins into repeatable implementations.
  • Identify problems, propose solutions, and drive them through completion.

About the company

Intent Lab builds Fleet, an autonomous AI system that acts as an engineering team to turn high-level intents into production-grade software, handling architecture, coding, deployment, testing, and ongoing optimization. Fleet takes natural-language requests and creates software tailored to a user’s constraints and workflow, instead of just generating code. It differentiates itself by delivering end-to-end lifecycle management with built-in quality and performance checks, resulting in production-ready systems. The company aims to shift how software is created by generating and refining software directly from user intent, producing fully verified solutions customized to each task.

Company Size

1-10

Company Stage

N/A

Total Funding

N/A

Headquarters

San Francisco, California

Founded

2026

Get referred to Intent Lab

See people who can refer or advise you

Simplify Jobs

Simplify's Take

What believers are saying

  • Intent Lab's August 2026 launch earned immediate attention from developers and AI infrastructure buyers.
  • The Fleet database demo passed all six million SQLite compatibility tests from scratch.
  • Job postings in May 2026 signal hiring, suggesting productization after the July launch.

What critics are saying

  • No public funding, revenue, or named customers are disclosed by September 7, 2026.
  • Cursor, Claude Code, and OpenAI Codex already own coding workflows, compressing Intent Lab's wedge.
  • NVIDIA and other platform vendors can ship similar autonomous agents, squeezing Intent Lab's margins fast.

What makes Intent Lab unique

  • Yangqing Jia launched Intent Lab in July 2026, bringing Caffe, ONNX, and PyTorch credibility.
  • Fleet targets end-to-end software delivery, not just code completion, verification, or refactoring.
  • Early demos showed 6.3x GLM-5.2 inference gains and six million SQLite tests passed.

Help us improve and share your feedback! Did you find this helpful?