legal_ops · finance · workflow
Coinbase builds enterprise AI agents: six-week sprint to standardize agentic automation for internal workflows
Coinbase internal teams ran manual, time-consuming, decision-heavy workflows in Institutional support, Onramp onboarding, and Listing legal review, with no standardized way to build, host, or audit AI agents at enterprise scale.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Manual work identified in automation domains
Work items in Institutional support, Onramp onboarding, and Listing legal review are identified as candidates for process automation.
Tools used
LangGraphLangChainLangSmithPython
Outcome
Two automations reached production with more than 25 hours saved per week; new agent build time fell from 12+ weeks to under a week, and more than half a dozen engineers were able to self-serve on the patterns.
What failed first
Low-code tools proved unsuitable for long-running operational flows because loading more tools and instructions into a prompt introduced context noise, making outputs harder to reproduce and individual steps harder to unit test or gate in CI.
Results
Time savedmore than 25 hours saved per week
Volumetwo
Grounding & classification
Source type: technical build writeup
30 fields verified against source quotes, 1 dropped as unverifiable.
agentic workflowai agentsummarizationpolicy documentbuilder submittedhuman review describedmetric backednamed customerproduction runtime claimedtools describedworkflow describedfinancial servicescycle time reductionemployee productivitytime savedtechnical build writeupback office opscustomer supportlegal opsagentic task executionai draft human approvalhuman review queue