XR Distributed Compute Engineer

Qualcomm•San Diego, CA

7d•$148,300 - $222,500•Onsite

About The Position

Multimodal, always on perception and agentic AI are driving designs and adoption of new personal, wearable, and eXtended Reality platforms. Qualcomm's XR Labs team is seeking a skilled systems engineer to design and build software systems for next generation of power-aware distributed agentic AI. In this role, you will design and implement the orchestration layer for a network of heterogeneous personal wireless devices—including wearables and companion platforms—that collaboratively sense and process complex multimodal inputs. Our goal is to move beyond simple prompt-response cycles toward autonomous, goal-oriented agent networks that can dynamically distribute compute workloads across local and edge resources. You will work at the intersection of high-performance system programming and cutting-edge AI, building the communication and execution backbone that allows these agents to discover one another, negotiate tasks, and synchronize shared contexts in real-time.

Requirements

Strong performance in modern, performant C++ with a deep understanding of multi-threading, memory management, and systems-level programming. (5+ years)
Hands-on experience building and deploying multi-agent networks using LangChain, AutoGen, or similar orchestration frameworks. (3+ years)
Proven track record in designing workload distribution logic, task scheduling, state synchronization, and handling of heterogeneous compute profiles. (3+ years)
Strong understanding of discovery protocols and advertising mechanisms.
Ability to work in a fast-paced R&D environment, bridging the gap between research and production-grade system implementation utilizing problem-solving skills.

Nice To Haves

Hands-on experience with WWAN, WiFi Aware/NAN and/or BT protocols for P2P ranging and communication.

Responsibilities

Develop and maintain performant services for agent orchestration, task decomposition, and workload distribution across a mesh of personal devices.
Implement robust device-to-device (A2A) advertising and discovery mechanisms, ensuring low-latency handoffs and stable peer-to-peer connectivity in dynamic environments.
Coordinate the processing of multimodal sensing data across distributed nodes, optimizing for power-aware, latency-optimized inference with minimal communication overhead.
Utilize agentic patterns and frameworks to manage complex reasoning task routing, tool invocation, and context propagation between specialized agents.
Build, optimize, and benchmark distributed compute workloads to meet the stringent latency and thermal constraints of wearable hardware.