Responsibilities
- Design multi-agent architectures with advanced orchestration tools to fully automate intricate customer service workflows.
- Create and maintain scalable integrations via Model Context Protocol to link language models with internal APIs, data stores, and external platforms.
- Construct automated evaluation systems using LLMs as evaluators to track, assess, and enhance AI agent performance in live environments.
- Apply retrieval-augmented generation, model fine-tuning, and prompt optimization to ensure chatbot outputs are factually grounded, precise, and aligned with compliance standards.
- Develop robust, low-latency Python-based services capable of managing high-volume traffic across asynchronous microservice ecosystems.
- Collaborate with design and product teams to convert sophisticated business rules into functional agent behaviors within a decentralized system.
Work Arrangement
Hybrid
Other
- Operates with a remote-first culture but allows for non-remote roles.
- Requires periodic in-person gatherings every quarter known as 'surges' for focused collaboration.
- Applicants may apply to no more than three positions within any six-month window.
- Committed to equal opportunity employment regardless of background.
- Individuals with criminal records will be evaluated in accordance with applicable laws.
- U.S.-based applicants should review Employee Rights, Know Your Rights, and E-Verify Participation Notice.
- Accessibility accommodations are available by emailing accommodations[at]coinbase.com.
- Submission of an application constitutes agreement to the Candidate Privacy Notice.
- U.S. applicants agree to dispute arbitration by submitting their application.
