Senior AI Engineer — Permitting Intelligence at Scale

Pulley; San Francisco (CA); Hybrid; USD 230,000 - 280,000; Full time

Benefits offered by this job

Competitive salary

Equity

Health insurance

Flexible PTO

Remote work options

401(k)

Job summary

Pulley is building AI-powered permitting infrastructure in the SF Bay Area and beyond. This senior AI engineer role focuses on turning messy permitting inputs into fast, structured outputs with end-to-end ownership—from data intake to production features and monitoring.

You will shape the product surface with LLMs, work on evaluation and observability, and mentor teammates while delivering reliable, customer-facing capabilities. Remote options exist in coordination with in-person days.

Qualifications

• 4+ years of software engineering experience with a focus on production ML/LLM systems.
• End-to-end ownership of an LLM-powered product surface, from requirements to production.
• Deep hands-on experience with large language models in production.
• Experience designing evals to ensure reliability and quality of features.
• Experience building AI coding agents and shipping agent-driven workflows.
• Based in the SF Bay Area and willing to work in person 4 days a week.

Responsibilities

• Own AI-powered features end-to-end—from user interactions to production deployment.
• Turn unstructured permitting documents and regulations into structured outputs via extraction and classification.
• Build evaluation and observability foundations to measure quality and detect regressions.
• Work with AI agents as a daily practice, directing and shipping agent-driven work.
• Make technical and product decisions that directly impact customers and projects.
• Mentor engineers and raise the bar for quality and velocity.

Skills

Software engineering

LLM in production

Product mindset

Eval design

Tools

TypeScript

React

Google Cloud

Python

Job description

Pulley is building AI-powered permitting infrastructure in the SF Bay Area and beyond. This senior AI engineer role focuses on turning messy permitting inputs into fast, structured outputs with end-to-end ownership—from data intake to production features and monitoring.

You will shape the product surface with LLMs, work on evaluation and observability, and mentor teammates while delivering reliable, customer-facing capabilities. Remote options exist in coordination with in-person days.