AI Engineer: Agentic Systems & LLM Architect

Bison; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Benefits offered by this job

Medical, dental, vision coverage

401(k) plan

Relocation support

Snacks and meals in office

Unlimited PTO

Job summary

Zyphra, an artificial intelligence company based in San Francisco, is seeking an AI Engineer to help build next-generation agentic systems that autonomously navigate the web, interact with filesystems, and complete complex user tasks across desktop and browser environments.

You will work on frontend interfaces, secure sandboxes, large-scale document search, and integration of LLMs and vision models, contributing to production-grade ML systems with a bias toward speed and curiosity.

Qualifications

• Proficiency in Python and building ML-driven apps.
• Experience with Windows/macOS APIs for screen reading and file interaction.
• Experience developing browser extensions or automation tools with browser control.
• Understanding of LLMs, prompting, and orchestration for multi-step reasoning.
• Ability to work across the ML stack from model to serving infrastructure.
• Experience designing or working with secure/virtualized execution environments.
• Excellent communication across product, research, and engineering teams.

Responsibilities

• Design agentic system that interacts with browsers, OS, and filesystems.
• Build search and retrieval pipelines for large-scale data.
• Integrate LLMs, vision models, RL, and scaffolding for autonomous decisions.
• Engineer secure virtualized runtimes and backend services.
• Deliver production-grade ML systems with fast iteration and collaboration.

Skills

Python

Desktop OS APIs

Browser automation

LLMs & prompts

Full ML stack

Secure execution envs

UI/Agent design

Model fine-tuning

Tools

Puppeteer

Playwright

Job description

Zyphra, an artificial intelligence company based in San Francisco, is seeking an AI Engineer to help build next-generation agentic systems that autonomously navigate the web, interact with filesystems, and complete complex user tasks across desktop and browser environments.

You will work on frontend interfaces, secure sandboxes, large-scale document search, and integration of LLMs and vision models, contributing to production-grade ML systems with a bias toward speed and curiosity.