Real-time Collaboration Engineer

Migaru AI builds an AI-native CRM that keeps itself current by reading a sales team's email, calendar, and calls, filing conversations against the right accounts and deals so reps spend their time selling instead of updating records. We're looking for a Real-time Collaboration Engineer to help build the infrastructure that keeps records, drafts, and team views in sync as multiple people and AI agents read and write to the same data at once.

What you'll do

• Design and build real-time sync infrastructure so CRM records, drafts, and dashboards update live across users and sessions
• Implement conflict resolution and concurrency strategies (e.g. CRDTs, operational transforms, or similar) for shared, frequently-updated records
• Build and maintain the systems that let human review and AI-generated updates merge safely without data loss or race conditions
• Optimize for low-latency updates across web clients, background agents, and integrations with external tools
• Work with backend and product teams to extend real-time collaboration to new surfaces, such as shared pipelines, review queues, and live dashboards
• Instrument, test, and debug distributed state issues, including edge cases around offline/reconnect behavior and out-of-order updates

What we're looking for

• Experience building real-time, multi-user systems in production (collaborative editors, live dashboards, multiplayer apps, or similar)
• Strong understanding of concurrency, state synchronization, and conflict resolution techniques (CRDTs, OT, or comparable approaches)
• Solid backend engineering skills, comfortable designing APIs and data models that support live, concurrent updates
• Experience with WebSockets or similar real-time transport layers, and the tradeoffs involved in scaling them
• Ability to reason carefully about correctness and edge cases in distributed, asynchronous systems
• Comfortable working closely with product and other engineers to translate collaboration requirements into reliable systems

Nice to have

• Experience with CRDT libraries (e.g. Yjs, Automerge) or building your own
• Background working on CRM, productivity, or collaborative document tools
• Experience integrating AI-generated content into human review/approval workflows