Software Engineer

Cotidal; Palo Alto (CA); On-site; USD 180,000 - 230,000; Full time

Benefits offered by this job

Competitive salary

Equity

Relocation support

Visa sponsorship

Job summary

Cotidal is assembling a software and systems engineering team to build a compute platform for AI workloads. You will contribute across accelerator clusters, distributed infrastructure, training/inference systems, and customer-facing products.

You’ll join a small team, work across boundaries, and help translate complex problems into reliable software and scalable systems, collaborating with teammates and customers throughout development.

Qualifications

• Experience building reliable software systems and shipped products.
• Ability to dive into the stack, runtime, or hardware to identify bottlenecks and fix them.
• Proven capability to translate user needs into usable software features.

Responsibilities

• Build and operate accelerator clusters powering AI workloads.
• Develop the distributed control plane for the compute platform.
• Improve training/inference frameworks and model serving.
• Collaborate on product-facing APIs and backend services.

Tools

Kubernetes controllers

Distributed systems

PyTorch

JAX

XLA

vLLM

SGLang

Job description

Who We Are

Supercomputing built for AI.

Cotidal is building the computing infrastructure that AI companies and research labs need to train and run their models. Starting with TPUs, we design the hardware and software together around their workloads, with the goal of making powerful AI chips easier to use at scale.

Our mission is endless compute for humanity. Over time, we want customers to choose across chip stacks without rebuilding their infrastructure.

The Role

We’re looking for software and systems engineers to help build Cotidal’s compute platform. Our work spans accelerator clusters, distributed infrastructure, training and inference systems, and customer products.

You’ll join a small engineering team building these systems from their earliest deployments through production scale. We welcome engineers with deep experience in one area who enjoy learning unfamiliar systems and working across boundaries. You’ll work closely with teammates and customers to understand problems, make technical decisions, and carry projects through to working software and reliable systems.

What You Will Do

You’ll help shape the projects you work on, drawing on your strengths, experience, and interests to tackle the problems ahead. The areas below offer examples of where you could contribute, with room to work across them. We don’t expect experience in all of them.

HPC & Cluster Systems: Build and operate the accelerator clusters that power demanding AI workloads. Work across Linux, networking, storage, and hardware health to bring systems online, diagnose failures, and keep large fleets reliable and efficient.

Software Infrastructure & Control Plane: Build the distributed systems that turn individual machines into a compute platform. Work on provisioning, scheduling, resource management, and orchestration, making complex infrastructure reliable as the fleet and its workloads grow.

Training, Inference & Performance: Help AI teams get more from their compute. Work on training and inference frameworks, model serving, and performance across accelerators — from understanding a workload’s bottlenecks to improving how it runs in production.

Product Engineering: Build the products through which customers use Cotidal. Work across interfaces, APIs, and backend systems, turning customer needs into useful software and owning features from the first conversation through production.

What We Are Looking For

• You’ve built things people rely on—and been there when they broke, outgrew the original design, or got used in ways you never expected.

• You follow a problem wherever it leads: dive deep into the stack, runtime, or hardware to understand a bottleneck, then zoom out to see whether fixing it solves the bigger problem.

• When the path isn’t clear, you build something to find out—a prototype, a benchmark, a small working slice. You use what you learn to decide what deserves a bigger investment.

• You have opinions about how systems should be built, can explain the tradeoffs, and have changed your mind when the evidence called for it.

Especially Valuable

Experience in one or more of the following:

• Kubernetes controllers, schedulers, or distributed systems handling resource allocation, tenant isolation, and failure recovery.

• Bare-metal provisioning and hardware bring-up; RDMA, collective communication, or distributed storage.

• Training or inference optimization with PyTorch, JAX, XLA, vLLM, or SGLang—especially batching, KV caches, quantization, and parallelism.

• Building complex, user-facing products across the full stack, with strong product judgment and attention to interaction design, performance, and reliability.

• Hands-on experience with GPU, TPUs, Trainium, Instinct, Jalapeño, WSE, or other accelerator architectures.

Why Join Cotidal

Work across cluster infrastructure, fleet software, AI workload performance and customer-facing APIs and workflows. Agree the starting problem and ownership boundary with the team.

Own work from design through delivery, working directly with the founders. Help shape the foundations of an early company through direct discussion and work across disciplines.

Location & What We Offer

Location: Palo Alto, in person.

Compensation: Competitive salary and equity.

Visa sponsorship: We sponsor work visas and help you navigate the process.

Benefits: Medical, dental, and vision coverage, unlimited PTO, and relocation support as needed.