Senior ML Performance Engineer

About the role

Atlassian is hiring a Senior ML Performance Engineer for its Agentic AI Integration team, which builds the Rovo MCP Server, agent-to-agent integrations and AI-powered features such as Claude Skills and ChatGPT/Claude Apps. The role works on AI integrations into the Atlassian ecosystem, from prototyping to scalable enterprise products, and is open to Mountain View, San Francisco, Seattle or remote.

Key responsibilities

• Tune MCP and agent-facing servers for latency, reliability, token efficiency and tool-selection quality, including dynamic tool discovery, context and response optimization, observability, automated evals and semantic retrieval.
• Design, build and evolve MCP servers, tools and agent-facing APIs with clear, safe contracts.
• Build accessible, responsive React and TypeScript experiences and reusable components for agent capabilities and MCP tools.
• Integrate GraphQL, REST, SDKs, streaming responses and real-time data into AI workflows.
• Optimize token and context efficiency through caching, pagination, selective field retrieval and reduced tool-call loops.
• Improve performance and reliability across front-end clients, gateways, search services and downstream systems via observability, tracing, SLOs and production diagnostics.
• Build semantic retrieval with embeddings, vector indexes, hybrid search, ranking and reranking.
• Run AI/ML quality programs with JTBD-based evaluations, hallucination and bias detection, safety testing and human feedback, integrated into CI/CD.
• Deliver secure AI integrations with OAuth 2.1, tenant isolation, audit logging and prompt-injection defenses.

Required profile

• 7+ years of software engineering experience building and operating enterprise systems, APIs or cloud-native products.
• Experience measuring latency, throughput, error rates, availability, token consumption and end-to-end task cost.
• Ability to lead cross-functional initiatives, communicate with non-technical partners and mentor engineers.

Required skills

• TypeScript/JavaScript and React; hands-on experience with MCP servers, agent APIs or similar frameworks.
• Distributed systems principles: concurrency, caching, retries, timeouts, autoscaling, load shedding.
• AI/ML evaluation and quality engineering, including benchmark design and regression prevention.
• GraphQL, REST, JSON Schema, streaming APIs and event-driven systems.
• At least one additional systems language such as Python or Go.
• Preferred: MCP architecture, A2A specification, observability, vector databases, API governance or open-source AI contributions.

What we offer

• Base pay for new hires: Zone A $180,000-$235,000, Zone B $162,000-$211,500, Zone C $149,400-$195,050.
• Possible eligibility for benefits, bonuses, commissions and equity.
• Flexible work arrangements and virtual interviews and onboarding.