Staff-level engineer building and operating Coinbase's LLM/agent infrastructure including gateway, runtime, knowledge bases, and applied AI agents. Own multi-quarter platform initiatives and partner with frontier labs.
Salary not listed
Remote7+ YOEML Engineering
About the role
What you'll do
LLM Gateway: one API for 60+ models across all major providers, including self hosted open weight models, with auth, PII redaction, quotas, fallbacks, and cost optimization
AI Hub: single front door for agents, skills, MCP servers, knowledge bases, models, memories, and keys
Agent and MCP runtime: microVM sandboxes per agent run, a governed MCP gateway over 125+ servers, cross trust agent identity, tier 1 agents for code, analytics, and exec personas
Knowledge Bases: connect any data source, auto provision vector and markdown stores, queryable by any agent over MCP
AI Spend Hub: FinOps for AI, spend attribution and governance across the company
Evaluation and observability: in house LLM as judge harness with full tracing and eval tooling
Company wide chat and low code workflow surfaces, thousands of weekly actives each
Applied AI: embed with teams to build production agents, build the feedback loops that let subject matter experts refine them, and apply traditional ML including fine tuning where it beats prompting
Utilize and build generative AI responsibly, maintaining human oversight to deliver business-ready outputs and drive measurable improvements in workflow efficiency, cost, and quality
Required Skills and Experience
Own multi quarter, high impact initiatives on the platform (gateway, runtime, knowledge, governance) or the applied agents on top
Partner across engineering, security, legal, finance, and product
Work directly with partners at the frontier labs and major cloud providers
Build and optimize the RL training framework and infrastructure for large-scale workloads at SpaceXAI, from ablations to production runs. Requires experience with distributed systems and proficiency in Python, JAX, Rust, or C++.
180k – 440k
On-site5+ YOEML Engineering
Staff Software Engineer, AI & App Experience
ChimeNew York, NY +1
Staff Software Engineer building and scaling Jade, Chime's LLM-powered AI financial assistant. Sets technical direction for agent systems, evals, and guardrails while staying hands-on with prototyping, backend services, and cross-functional product delivery. Requires 8+ years production experience, deep AI/LLM fluency, and technical leadership.
Build and own validation pipelines, CI/CD infrastructure, and platform integrations to launch frontier models and inference features reliably across AWS, GCP, and Azure. Requires strong large-scale distributed systems experience and track record improving release velocity.
320k – 485k
Hybrid7+ YOEML Engineering
Senior Staff Software Engineer, Legal Automation
CoinbaseUnited States
Senior Staff Software Engineer building an AI agent platform and automated workflows to transform Coinbase's Legal organization. Architect production-grade LLM and multi-agent systems that replace manual legal processes such as agreement redlining and governance.
254k – 299k
Remote12+ YOEML Engineering
Staff Machine Learning Engineer, Computer Vision
PinterestSan Francisco, CA
Staff Machine Learning Engineer developing state-of-the-art visual encoders and multimodal models at Pinterest Labs. Prototype visual reasoning tools, train billion-scale models on rich visual-text data, ship to production for recommender systems and VLMs, publish research, and mentor juniors. Requires strong CV/ML background, publications, and PhD or equivalent.