Principal Software Engineer, AI

PermitFlow•New York, NY
•$250,000 - $300,000•Hybrid

About The Position

PermitFlow is hiring a Staff- or Principal-level engineer to build its core AI Assistant, AI infrastructure, and workflow agents for construction permitting and compliance. The role combines product and technical expertise to deliver high-quality agents in a complex domain and advance PermitFlow’s agent infrastructure and data flywheel. The successful candidate will bring a record of product delivery and hands-on experience building AI and agentic infrastructure from the ground up, ship quickly, instrument systems, maintain a high quality bar, and stay close to users. The position reports directly to the VP of Engineering. Level is determined based on past experience and interview performance. The role focuses on building an assistant that helps customers understand construction documents, extract project details, confirm requirements, and complete forms, while reducing submission errors and delays. PermitFlow is headquartered in New York City and uses a hybrid schedule with three in-office days per week. NYC-based candidates or candidates open to relocation are preferred.

Requirements

  • 10 or more years of software engineering experience, including recent experience building LLM-powered products.
  • Experience shipping an assistant or agentic product at meaningful scale.
  • Deep, current knowledge of the agentic stack, including tool calling and design, context engineering, memory, retrieval, and orchestration patterns.
  • Experience building evaluation harnesses that influenced a team’s priorities and ability to explain what makes an evaluation trustworthy.
  • Strong software engineering skills, including ownership of production services and writing maintainable code.

Nice To Haves

  • Preference for candidates based in New York City or open to relocation.

Responsibilities

  • Own the agent architecture, including the reasoning loop, planning and decomposition strategy, state and session model, and recovery behavior for long-running tasks.
  • Build the context layer, including retrieval, memory, and context-window management across project documents, jurisdictional requirements, prior submissions, and customer-specific conventions.
  • Design the assistant’s action surface, including callable tools, permissions, and human-in-the-loop checkpoints for expensive or irreversible actions.
  • Build offline and online evaluation harnesses to measure task completion and failure modes, and establish evidence-based iteration loops to improve agent quality.
  • Establish abstractions, SDKs, and review standards for other engineers building agentic features.
  • Work directly with permit coordinators and builders, observe the assistant in real-world conditions, and apply those findings to system design.

Benefits

  • Competitive salary and meaningful equity in a high-growth company.
  • 100% company-paid base medical, dental, and vision coverage for employees.
  • Healthcare FSA.
  • 401(k) savings plan.
  • Unlimited PTO and paid family leave.
  • Home office and equipment stipend.
  • Daily in-office lunch and dinner.
  • Commuter benefits for pre-tax transit and parking.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service