Software Engineer, Agentic Runtime

About this role

Glean seeks a Software Engineer to build the core runtime infrastructure powering AI agents and assistants at enterprise scale. You'll design low-latency, secure services for orchestration, tool calling, and streaming responses while optimizing performance and reliability across distributed systems.

What you'll do

• Own end-to-end runtime problems from architecture and design through production launch and reliability
• Build core services for session lifecycle, streaming (gRPC/WebSockets), tool execution, and memory management
• Optimize latency (p50/p95), improve tail behavior, and reduce token/tool costs
• Integrate with LLM providers (OpenAI, Anthropic, Google Gemini) and evaluation frameworks
• Harden platform with fault isolation, retries, circuit-breaking, and graceful degradation
• Establish observability (tracing, metrics, logs) and on-call playbooks with SLOs

What they're looking for

• Distributed systems design
• Low-latency service architecture
• gRPC and WebSocket streaming
• Production observability and monitoring
• Fault tolerance and reliability patterns
• LLM API integrations
• Performance optimization
• Backend software engineering