The Lead Software Engineer - AI Agents will be the primary architect and technical lead for the team building all AI agents across the ecosystem. Acting as a hands-on leader and Technical Lead, you will be responsible for the team's overall delivery, translating complex product requirements into actionable technical tasks for a small squad. You will design and build high-stakes agentic architectures that power our multi-sided marketplace—including buyer and seller-side agents, platform onboarding assistants, and automated maintenance agents—while continuing to reason across multi-modal data sources like OEM manuals, parts catalogs, and repair history. You will remain deeply embedded in the codebase to solve the "grounding problem" where accuracy is non-negotiable.
Responsibilities
Technical Execution: Lead the technical delivery by decomposing high-level roadmaps into granular, actionable tasks for the agent development team. You are the execution owner responsible for mentoring engineers and ensuring high-quality output through rigorous architectural oversight.
Agentic Architecture Design: Architect and implement advanced agentic systems and RAG pipelines that perform multi-step reasoning across disparate data sources to answer complex diagnostic questions. This includes expanding beyond diagnostic systems to develop intelligent buyer/seller marketplace agents and platform agents that automate customer onboarding and ecosystem maintenance.
Solving the Grounding Problem: Lead the development of confidence scoring systems and validation workflows to ensure zero tolerance for plausible-but-wrong AI output in heavy equipment service.
Multimodal Agent Development: Design and code agents capable of processing and extracting procedures from diverse inputs, including text, images, and video, to generate verified training content.
Customer AI Integration Strategy: Architect the APIs and data access patterns that allow customers to build their own AI solutions using company data while maintaining strict hallucination and scope controls.
Production AI Scaling: Design and deploy production-grade AI capabilities within Google Cloud Platform and AWS, ensuring systems scale reliably with customer demand while managing compute costs effectively.
Evaluation Frameworks: Establish the metrics and gold dataset pipelines needed to measure retrieval quality and improve agent accuracy over time.
AI Native SDLC Actively utilize autonomous coding agents to plan, generate, and test AI logic, maintaining high velocity while ensuring code quality through agent-assisted peer reviews.
Skills and Experience
10 years of Software Engineering experience, with a proven history of shipping complex, distributed systems at scale.
Proven track record of designing and deploying agentic architectures, multi-step reasoning systems, and complex RAG pipelines in production.
Technical Stack: Mastery of Python (FastAPI, Django) and modern Agent development/observability frameworks (e.g., LangGraph, LangChain, LangSmith, Google ADK). Solid hands-on experience with foundational AI/ML and MLOps frameworks (e.g., PyTorch, Scikit-learn, MLflow, Weights & Biases).
AI & Retrieval Mastery: Deep experience in Agentic Architectures, GraphRAG, and multi-modal agent design (processing images/video alongside text), Agent Evaluation.
Cloud Infrastructure: Expertise in deploying AI workloads at scale across Google Cloud Platform and AWS
Coding Agents: Demonstrated proficiency in using coding agents to accelerate the SDLC and plan and code complex engineering tasks.
