Staff Software Engineer, AI Platform

Our Agent Studio team is building the foundational platform that enables every engineering team across the company to create, deploy, and evaluate AI agents. We're looking for a Platform Engineer with a forward-deployed mindset—someone who can architect and ship robust platform capabilities, while also rolling up their sleeves alongside product teams to ensure those capabilities succeed in production.You won't build the platform in isolation. You will collaborate closely with consuming teams, understand their agent use cases firsthand, co-build tools and integrations, and translate those learnings back into a scalable, self-serve platform.This position reports to a Senior Manager, Software Engineering and will be 2 days per week hybrid role in our Austin office. We’re looking for someone to join us immediately.What You'll Do:Forward Deployed / Team Enablement (50%)Co-Building & Enablement: Partner directly with product teams to unblock agent use cases, co-build initial agents, and drive adoption via office hours, integration guides, and reference implementations.Domain & Persona Alignment: Leverage knowledge of the built environment and ConTech platforms (Procore, Autodesk CC, Buildertrend, OpenSpace) to address key friction points for GCs, Subcontractors, Owners, PMs, and Field Engineers.Platform Feedback Loop: Translate fragmented, team-specific requirements into generalized platform capabilities and concrete roadmap proposals.Technology Evaluation: Collaborate with the AI Architect and engineering leads to evaluate, benchmark, and adopt emerging agent frameworks and tools.Platform Engineering (50%)Runtime & Security Infrastructure: Design and build the agent runtime, tool registry, and execution security controls, including process isolation, egress restrictions, and credential scoping.Evaluation Frameworks: Build automated testing harnesses, LLM-as-a-judge pipelines, regression tracking, and quality dashboards for agents.Agent Observability: Own end-to-end tracing using OpenTelemetry, session replay, cost attribution, and latency profiling across multi-step agent runs.MCP, Integrations & API Contracts: Maintain Model Context Protocol (MCP) servers, third-party tool integrations, and stable, well-versioned platform API schemas for cross-org consumption.What We're Looking For:5+ years of software engineering experience, with at least 2 years focused on developer platforms, internal tooling, or infrastructure.Hands-on experience building and deploying production LLM applications or agentic systems.Deep proficiency in Python and/or TypeScript; comfortable reading and navigating Go codebases.Proven track record designing and maintaining developer-facing REST or GraphQL APIs.Strong grasp of agent architecture: tool usage, multi-step reasoning, memory management, structured output, and error handling.Pragmatic "bias for action"—capable of shipping rapidly without compromising long-term platform integrity.Familiarity with the ConTech landscape (Procore, Autodesk Construction Cloud, Buildertrend, OpenSpace).Clear understanding of core customer personas in the built environment and their operational pain points.Preferred QualificationsBackground in ConTech support, specifically troubleshooting AI features, LLM interfaces, or automated toolsets.Hands-on experience with Model Context Protocol (MCP) or modern tool-calling frameworks.Familiarity with LLM evaluation methodologies (automated judging, red-teaming, eval-driven development).Experience with workflow orchestration engines (e.g., Temporal, Prefect, Airflow) for long-running agent tasks.Experience implementing AI/ML observability tools (e.g., OpenTelemetry, Langfuse, Datadog).Prior experience in forward-deployed, solutions engineering, or developer advocacy roles.Additional InformationBase Pay Range:168,560.00 - 231,770.00 USD AnnualThis role may also be eligible for Equity Compensation and/or Bonus Incentive Compensation. Procore is committed to offering competitive, fair, and commensurate compensation. Actual compensation will be based on a candidate’s job-related skills, experience, education or training, and location.For Los Angeles County (unincorporated) Candidates:Procore will consider for employment all qualified applicants, including those with arrest or conviction records, in accordance with the requirements of applicable federal, state, and local laws, including the City of Los Angeles’ Fair Chance Initiative for Hiring Ordinance, the Los Angeles County Fair Chance Ordinance for Employers, and the California Fair Chance Act.A criminal history may have a direct, adverse, and negative relationship on the following job duties, potentially resulting in the withdrawal of the conditional offer of employment: 1. appropriately managing, accessing, and handling confidential information including proprietary and trade secret information, as well as accessing Procore's information technology systems and platforms; 2. interacting with and occasionally having unsupervised contact with internal/external customers, stakeholders, and/or colleagues; and 3. exercising sound judgment.SummaryLocation: US - Texas - AustinType: Full time