Callosum develops system-level software that allows different AI chip architectures to interoperate, enabling flexible, sovereign AI infrastructure.
ABOUT US The last era of AI scaled on a single bet: bigger models, more identical chips, more data. As problems grow more complex and the requirements of intelligence more diverse, that bet is breaking down. Real-world problems are heterogeneous: no single model or chip can solve them alone. The next era of AI requires heterogeneity at the infrastructure level - diverse models on diverse chips, each with distinct strengths, co-evolving into systems of capability that move the Pareto frontier of what is possible. That's what we are building. Callosum is the Intelligent Systems Company. We started from questioning what actually creates intelligence. We believe there is no single answer, but rather a system-level solution. We co-evolve models, workflows, and silicon together to show that intelligence does not come from a single component, but it emerges from the diversity of co-optimised mechanisms working together and aware of each other. Heterogeneity will define the next era of compute, and is a principle that holds in biological, neuronal, and economic systems alike. In early 2026 we launched with results showing orders of magnitude improvements in performance, and this is only the beginning. Agentic AI is the future of how intelligence is deployed: multi-step, long-horizon, and operating in changing environments. These systems are inherently heterogeneous, and can only be as powerful as the infrastructure that runs them. We are engineers and scientists based in London, working together across the full depth of the stack. We are curious, intellectually honest, and building what doesn't exist yet. If you thrive on uncharted territory and are energised by the scale of the challenge, we'd love to hear from you. ABOUT THE ROLE You'll own the architecture and the technical bar for the API platform our customers integrate against. A customer sends a request for a workflow. Your platform authenticates it, routes it to the right execution backend, scales to meet the load, and keeps every customer's traffic and data fully separate from every other's. The model execution itself sits behind endpoints your platform consumes, so this is a role about the API and orchestration layer, not the serving internals. This is the most senior engineering hire on the platform. It's a build role first and a leadership role second. You'll write the core systems yourself for some time before you mostly direct others. You'll set the foundation the rest of the engineering team builds on, and grow a small platform team around you, starting with a reliability engineer and a product engineer. WHAT YOU'LL BUILD - The API surface customers integrate against, and the orchestration layer behind it: request lifecycle, routing, and state management - DNS routing, load balancing, and autoscaling, so the platform meets demand without anyone touching it by hand - Multi-tenancy and isolation - Per-workflow scaling and deployment - The technical bar. Design standards, code review, and the architectural decisions that are expensive to reverse later - Performance and scalability as customer volume grows to millions of concurrent requests - Production reliability, as a partnership. You set the architecture that makes reliability achievable. A dedicated reliability engineer owns day-to-day operations WHAT YOU BRING - Deep experience designing, building, and operating production API platforms that customers depend on - Strong distributed-systems fundamentals: stateless and stateful services, routing, queueing, autoscaling, multi-tenancy, fault tolerance - A track record of owning architecture for a system at scale - The judgment to know which architectural decisions are reversible and which are not - Hands-on, with a bias for action. You want to build, not just direct, and an early-stage environment where you set the foundations is what you're after - Comfort with ambiguity and a strong sense of ownership. You'll make consequential calls without complete information WHAT SETS YOU APART - Experience with multi-tenant systems where data separation between customers is a hard requirement - Familiarity with high-throughput inference or agent workloads, and a feel for how different workload types place different demands on the infrastructure around them - Experience running production API infrastructure that meets high-assurance standards - Open-source contributions to relevant infrastructure, or production systems whose scale and complexity you can speak to in detail - Early-stage and AI-native company experience WHAT WE OFFER - Competitive Salary, determined by skills and experience - Equity & Ownership - Private healthcare - We offer Visa sponsorship and relocation benefits to hire the best in the world - We work in person at our London office. You'll have the tools, space and setup to do your best work, and if you have specific needs, just tell us We're committed to building an inclusive workplace where everyone feels welcome, and believe in equal opportunities for all.