Staff AI Engineer

Location New York City Employment Type Full time Department Product Design + Engineering Software Engineering Why this role matters We're on a mission to automate one of the most costly and expertise-dependent bottlenecks in the built environment — construction plan review . Today, plan review is slow, expensive, and highly manual, requiring licensed experts to navigate thousands of unique jurisdictional construction codes and complex architectural documents. We believe AI can help. As a key hire in our AI engineering organization, you’ll operate with a founder mindset, help in defining our long‑term data strategy, and help to grow multiple squads behind it. GreenLite is building agentic workflows that compress permitting cycle time across thousands of jurisdictions—spanning permit intake, checklist generation, document QA, code citation, and review ops inside LiteTable. Your work turns messy PDFs/CAD plans and scattered code texts into decisive, auditable agent actions that help customers get permits faster. Our agent stack leans on: Bedrock AgentCore to deploy and operate agents securely at scale with session isolation, long-running workloads, built-in tools, memory, identity, gateway, and observability. LangGraph for graph-based orchestration and error‑tolerant control flow. Strands Agents for structured reasoning and tool use. What you’ll do: Design & ship production agents: Own one or more high‑impact agent workflows (e.g., Permit Intake Triage, Smart Document QA, Compliance Comment Copilot, Code Lookup inside LiteTable). Compose multi‑step graphs (LangGraph) with Strands‑based reasoning/tooling and develop internal/external tools for targeting heterogeneous datasets. Operationalize on Bedrock AgentCore: Use Runtime for secure, scalable hosting and streaming; Gateway to expose APIs as agent‑ready tools; Memory for persistent context; Identity for least‑privilege access; Observability for traces/metrics; and built‑in Browser/Code‑Interpreter where appropriate. Evaluation & safety harness: Stand up task‑level and end‑to‑end evals (success rate, cost, latency, human‑handoff rate) and regression suites; borrow ideas from AgentBench/WebArena/SWE‑bench where useful but bias toward domain‑grounded tests for permitting. Retrieval & knowledge: Partner with Data to wire agents to building‑code knowledge (RAG), evaluate vector store options (incl. S3 Vectors preview, or Pinecone ) and set up durable knowledge interfaces the agents can trust. UX handoffs: Collaborate with Product/Design to craft agent‑first reviewer UX in LiteTable (great inline citations, diffs, and remediation suggestions). Quality & reliability: Build fault‑tolerant flows (retries/rollbacks/compensation), observability dashboards, and on‑call runbooks for agent incidents. Technical leadership: Create internal libraries/templates for agent patterns; mentor engineers; review designs with domain experts (architects/code officials). You may be a fit if you have: Agent frameworks: Depth with LangGraph (graph orchestration, state, recovery) and Strands Agents (structured tool reasoning); experience integrating MCP clients/servers. Production chops: 7–12+ years shipping high‑reliability backend or ML systems (Python/TypeScript), cloud infra (AWS), containers (ECS/EKS), CI/CD, IaC, and secure integration patterns. Retrieval & data: Hands‑on with RAG, document stores/vector DBs (bonus: knowledge of Pinecone or early exposure to S3 Vectors ), schema/versioning for code texts and comments. Evaluation mindset: Ability to design realistic, auditable testbeds; familiarity with agent benchmarks (and their pitfalls) and how to turn real user flows into acceptance tests. Safety & governance: Policy-driven guardrails (Grounding checks, topic/word filters), auditability, and human‑in‑the‑loop controls. Domain empathy: Curiosity for building codes, plan sets, and reviewer workflows; comfort working with messy PDFs/CAD and plan‑review UX. Bedrock AgentCore knowledge: Nice to have - comfort with Runtime (session isolation, 8‑hour sessions, large payloads), Memory, Gateway, Identity, Observability, and built‑in Browser/Code‑Interpreter. Nice‑to‑haves: CrewAI/AutoGen/LlamaIndex experience; Apryse or similar PDF tooling; geospatial data; prior work in AEC/reg‑tech. What success looks like 30 days: One production agent behind a feature flag with observability and a basic evaluation harness; measurable reduction in reviewer clicks/time on a scoped task in LiteTable. Hands on work with some aspect of our data to improve it for agentic work. 60 days: A small portfolio of cooperating agents (intake + doc QA + code lookup) in shadow/beta with Guardrails and human‑handoff paths; clear reliability SLOs. 90 days: Participation in developing a plan for agents driving ≥25% faster reviews on target flows; stable on‑call; cost/latency under thresholds; repeatable patterns/template repos useful for future agents. Why Join Us? Shape the data platform and labeling strategy that future hires will depend on. Be an early technical leader on a high‑impact AI product and trailblaze the first AI model for the building compliance industry. Work directly with architects, engineers, and domain experts who shape the system. Contribute to making the built world more efficient, sustainable, and safe. Solve one of the most under‑explored AI problems in the real world (with a huge market size). Operating principles in action Own It : You act like the CEO of our data layer—no broken pipelines on your watch. Solve Real Problems : Every schema, pipeline, and quality check exists to deliver ML‑ready data that shortens training cycles and lifts model performance—turning raw construction files into fuel for CV & LLM models. Full Send : You move with urgency, never at the expense of data integrity. Be Real : Radical candor about trade‑offs and technical debt keeps the team honest. If you’re ready to turn raw architectural drawings into the data engine that powers the first AI plan‑reviewer, we’d love to meet you. Thrive With GreenLite Competitive Compensation - Generous base salary & access to our Employee Equity Program, so you can grow with us. Performance‑Based Annual Bonuses - Rewards for high‑impact results and contributions that move the needle. Premium Health Coverage - Comprehensive medical, dental, and vision insurance for full‑time team members and their families. 401(k) Retirement Plan - Helping you invest in your future with smart saving options. Parental Leave - Generous parental leave for all parents to support your growing family. Wellness Support - Monthly Wellness Stipend and full access to Wellhub, Talkspace, & Teladoc for your physical and mental well‑being. Weekly Team Lunches - Enjoy catered lunches every week in our NYC office. Great food, better company. Company‑Wide Team All Hands - Held twice a year, fostering transparency, alignment, and inspiration. Team‑Building Events - Regular opportunities to connect, collaborate, and celebrate as a team. Unlimited PTO - Flexible time off so you can recharge, travel, or take care of life as needed. Hybrid Work Environment - Our team thrives on collaboration, so we’re in the office 4 days per week. In the summer, from Memorial Day to Labor Day, we switch to a 3-day in‑office schedule to give everyone extra flexibility. Equal Opportunity Statement GreenLite values people from all walks of life and professional backgrounds. We understand not everyone will meet all the above qualifications on day one. That's okay. If you’re passionate about the construction industry and want the opportunity to grow in your career, we encourage you to apply. GreenLite is an equal employment opportunity employer, committed to an inclusive workplace where we do not discriminate on the basis of race, sex, gender, national origin, religion, sexual orientation, gender identity, marital or familial status, age, ancestry, disability, genetic information, or any other characteristic protected by applicable laws. We believe in diversity and encourage any qualified individual to apply. #J-18808-Ljbffr 53 Stations