Senior AI Engineer

Pulley is hiring a Senior AI Engineer to help build AI-powered permitting intelligence, evaluation, and observability for LLM features. This role owns AI-driven product surfaces end-to-end, turning unstructured permitting documents into structured outputs and shipping agent-driven work that moves from prototype to production.

You will work in San Francisco, CA (onsite), collaborating closely with the team and shaping both the technical approach and the outcomes that matter for customers.

What you’ll do

• Own AI-powered features from user conversations and defining what “correct” means for permitting workflows through prompt and pipeline design, evals, deployment, and production iteration
• Convert unstructured permitting documents, city regulations, and jurisdiction workflows into structured, reliable outputs including extraction, classification, retrieval, and agentic workflows across document types that were never designed for machine reading
• Build an evaluation and observability foundation that enables confident shipping of LLM-powered capabilities by defining ground truth, measuring quality and regressions, and determining whether model changes are real improvements
• Use AI agents as a daily practice by directing, reviewing, and shipping agent-driven work at high velocity while maintaining a high quality bar
• Make technical and product decisions that affect customers and their projects
• Raise the engineering bar through design review, mentorship, and the standards you set in your own LLM builds

What you bring

• 4+ years of software engineering experience, with a substantial portion building production LLM or ML systems
• Experience owning an LLM-powered product surface end-to-end, including requirements through production and the unglamorous parts such as data quality, eval design, cost and latency, and failure handling
• Deep hands-on production experience with large language models: prompting, retrieval-augmented generation, structured extraction, tool use, and agentic workflows, plus the judgment to pick the right approach
• Proven ability to design evals and make LLM-powered features reliable in production
• Real experience building with AI coding agents, including shipping work where agents performed substantive implementation under your direction
• Ability to architect durable systems while making pragmatic tradeoffs
• Based in the San Francisco Bay Area and willing to work in person 4 days a week

Tools you’ll work with

• LLM, ML, retrieval-augmented generation
• TypeScript, React
• Google Cloud
• OCR and vision-language models

Compensation and employment

• Salary: USD $230,000 to $280,000 per year
• Equity: Offers Equity
• Employment type: Full time
• Department: Engineering

Nice to have

• Experience with document understanding at scale, including OCR, layout-aware parsing, or vision-language models for scanned PDFs, drawings, or forms
• Experience fine-tuning models or building data pipelines to produce training and eval sets from real-world usage
• Experience in construction tech, govtech, proptech, or similar domains where the challenge is messy documents and processes
• Modern full-stack development experience and willingness to work in application code that puts AI features in front of users (TypeScript, React, Google Cloud)
• Startup experience where you helped build the team, not only the product
• Experience mentoring engineers or leading technical direction across teams

What it’s like to work here

• Thrives in ambiguity by defining the right problem, then moving forward even when the path is unclear
• Product-minded approach, including talking to users to ensure the work solves customer problems
• Rigorous stance on “working,” prioritizing measurement through evals over demos
• Strong opinions on quality and velocity, focused on tools, abstractions, and processes that support both
• Ownership mindset when something is broken or missing