Member of Technical Staff - Agent Platform (Agent OS)
Core
Engineer and evolve the core Agent OS—a high-performance, resilient platform encompassing the dialog & policy engine, distributed context & memory, execution runtime, security isolation, voice runtime, and complex agentic orchestration frameworks.
Role type
Senior IC backend engineer (agentic systems & platform infrastructure)
Builds
A production-grade Agent OS underpinning all Boson agents, including low-code flows and advanced systems leveraging RAG, ReAct, and multi-step execution.
Domain
Enterprise AI, Agentic Systems, Distributed Systems
Deliverable
production ML models | product features | infrastructure
Required skills
backend engineering, distributed systems, agentic systems architecture, orchestration frameworks (LangChain/LlamaIndex), RAG, ReAct, multi-step planning, concurrent programming, security-sensitive system design, framework/library design, API design, observability (SLIs/SLOs), real-time media systems, low-latency optimization
Preferred skills
conversational AI platforms, real-time media (audio/video) systems, high-stakes enterprise platforms, federated learning, on-device personalization, adaptive policy systems
Technologies
LangChain, LlamaIndex, MCP, A2A, RAG, ReAct
Responsibilities
Define and implement the state machine for agent state representation and decision-making logic; Design and maintain high-performance context and memory systems with optimized RAG and vector retrieval pipelines; Architect robust agentic orchestration patterns including planner–executor schemes and multi-step workflows; Build and evolve internal SDKs/frameworks for composable graphs and execution chains; Own and optimize voice runtime components for streaming audio and low-latency protocols; Architect secure tooling and integration frameworks (MCP/A2A) with authz/authn and rate limiting; Define and monitor rigorous SLIs/SLOs for the Agent Platform; Design clear public-facing APIs and internal abstractions; Prototype future capabilities like on-device personalization and privacy-preserving federated learning.
Seniority
Senior, hands-on IC