Software Engineer, Machine Learning Infrastructure - Generative AI

About the Team DoorDash’s GenAI Platform team sits within Machine Learning Platform and builds the shared infrastructure that helps DoorDash, Wolt, and Deliveroo teams safely bring GenAI-powered products, agents, automation, and personalization to production. Our mission is to increase the velocity of business impact from GenAI. A central pillar of that work is our evaluation platform — the unified evals backbone that lets teams measure, trace, and trust the quality of LLM and agent systems across the company, powering trace/score ingestion, LLM-as-judge workflows, agent simulations, and LLM observability for the tens of millions of daily requests flowing through our LLM Gateway. We also own core platform surfaces including the Agent Gateway, open-weights model serving and batch inference, guardrails, and cost attribution. About the Role You will join a small, high-leverage team building production infrastructure for Generative AI at DoorDash, with a primary focus on our evals and LLM observability platform: the systems that let teams evaluate, trace, and continuously improve the quality of LLM and agent products. You’ll work across evaluation frameworks and SDKs, OpenTelemetry-based trace/score ingestion, LLM-as-judge and offline/online eval pipelines, agent simulations, data pipelines, backend services, and observability. This role is ideal for an engineer who enjoys building reliable measurement and quality primitives in a fast-moving technical area where product needs, model capabilities, vendor ecosystems, and evaluation methodologies are evolving quickly. You’re excited about this opportunity because you will… Build the infrastructure that helps DoorDash teams move GenAI ideas from prototype to production, increasing the velocity of business impact from AI across the company. Work on our unified evals platform — evaluation SDKs, OpenTelemetry trace/score ingestion, LLM-as-judge, offline and online eval pipelines, and agent simulations — alongside the LLM Gateway, Agent Gateway, open-weights model serving, guardrails, and cost attribution. Design scalable systems for evaluation workflows, trace/score ingestion, LLM observability, and agent simulation that power real customer and internal automation use cases Raise the quality bar for GenAI at DoorDash — giving product teams trustworthy, low-friction ways to measure model and agent quality, catch regressions, and compare across open-weight and closed-source models with observability and cost controls built in. Build platforms that support rapid experimentation while meeting production standards for latency, scale, monitoring, SLOs, playbooks, and operational excellence. Partner closely with ML engineers, product engineers, data scientists, and platform teams across DoorDash, Wolt, and Deliveroo to turn emerging GenAI capabilities into durable platform primitives. Shape the future of DoorDash’s centralized GenAI platform — closing the loop from evaluation and agent observability to agent optimization, where eval signals and traces drive automated evaluation, agent simulation, and post-training techniques (e.g., reward modeling and RLHF/RLVR evaluation) — enabling the next generation of AI-powered products, agents, automation, and personalization. We’re excited about you because… B.S., M.S., or PhD. in Computer Science or equivalent 3+ years of industry experience in software engineering Strong backend engineering fundamentals, especially in Python and distributed systems. Experience building production services, APIs, data pipelines, or ML infrastructure at scale. Experience operating systems in production, including observability, debugging, reliability, incident response, and performance/cost optimization. Hands-on experience with evaluation, LLM observability, or measurement systems for ML/LLM products in production — eval pipelines, tracing/scoring, offline/online quality metrics, or experimentation. Proficiency in using AI coding tools (e.g., Claude Code, Codex, Cursor) in the full software development lifecycle, including designing, generating code, testing, monitoring and releasing software Nice To Haves Depth in evaluation methodology — LLM-as-judge design and calibration, judge/eval drift detection, human-in-the-loop labeling, or eval harness design for agents and multi-step systems Experience with LLM observability and tracing (e.g., OpenTelemetry, trace/score ingestion) and building instrumentation SDKs Experience building and deploying AI agents or MCP servers in production, including agent evaluation or simulation Experience with data pipelines, streaming ingestion, and analytical stores (e.g., SQL, columnar/OLAP) for high-volume telemetry Experience with LLM gateways, model routing, vendor abstraction, or cost attribution Experience building developer platforms, internal platforms, or self-serve infrastructure Experience with Kubernetes, cloud infrastructure (AWS/GCP),