Heterogeneous Interconnects Architect for AI Systems

AI Startups UK; Greater London; Hybrid; GBP 90,000 - 150,000; Full time

Benefits offered by this job

Equity

Private healthcare

Visa sponsorship

London office

Job summary

Callosum in London is seeking an experienced systems engineer to architect and optimize the communication framework across heterogeneous accelerators. You will shape data paths, reduce latency, and enable seamless cross-device state movement.

You will work across RDMA paths, NICs, DPUs, and software layers, defining strategies to balance performance and cost while pushing the frontier of AI infrastructure in a hands-on, collaborative environment.

Qualifications

• Deep understanding of networking protocols and hardware.
• Experience with latency-sensitive systems and high-performance data movement.
• Ability to trace bottlenecks across software and hardware layers.

Responsibilities

• Characterise and simplify communication paths between accelerators, tracing latency and overhead across networking, runtime, memory, and device layers.
• Develop faster data movement across heterogeneous hardware, exploring RDMA, programmable NICs, DPUs, switches, and software approaches to cross-device memory management.
• Build flexible mechanisms for transferring model state across accelerator types and execution strategies, including KV cache connectors across different forms of parallelism.
• Evaluate existing and emerging scale-up and scale-out fabrics, informing cluster architecture and infrastructure integration.

Skills

Networking expertise

Systems performance

Latency optimization

Cross-disciplinary thinking

Tools

RDMA

Programmable NICs

DPUs

Interconnects

Job description

Callosum in London is seeking an experienced systems engineer to architect and optimize the communication framework across heterogeneous accelerators. You will shape data paths, reduce latency, and enable seamless cross-device state movement.

You will work across RDMA paths, NICs, DPUs, and software layers, defining strategies to balance performance and cost while pushing the frontier of AI infrastructure in a hands-on, collaborative environment.