Senior Full-Stack Engineer

Our product combines ML pipelines for deep neural network compression with small, security-critical SaaS services. That means one week you might be making a distributed training and compression pipeline reproducible and cost-efficient, and the next you might be designing an API, hardening a deployment, or building the UI that makes the whole thing usable.

We are not looking for someone who has used every tool in our stack. We are looking for engineers with enough architectural judgement and breadth to decide what should be built and why and enough fluency to review, debug, and take responsibility for the result.

Role Description

Productizing ML pipelines for DNN compression: turning experimental training, compression, and evaluation workflows into pipelines that are reproducible, observable, and cost-efficient to run continuously.

Designing and implementing backend services: Small, well-bounded services built to a high security standard from day one—primarily leveraging Python, but with room to evaluate and introduce the right tools for the job.

Building the frontends that expose our pipelines and services to users.

Owning deployment and operations of what you build: containers, Kubernetes, CI/CD, observability, and the security posture around it all.

Making architectural decisions with the team and writing them down clearly enough that both your colleagues and our coding agents can act on them.

Requirements

Minimum 7 years of experience building and running production software, with a track record of shipping systems you designed yourself.

Genuine breadth: you have worked on backends, data or ML pipelines, and infrastructure, and you are not afraid of the parts of the stack you know least well.

Solid grasp of software architecture and cloud platforms — you can describe a system's big picture, its boundaries, and its failure modes, and explain the trade-offs behind your choices.

Practical ML fluency: Demonstrated experience training, fine-tuning, and running inference with PyTorch or TensorFlow, alongside a clear grasp of modern ML tooling and pipeline architecture.

Proven ability to rigorously evaluate trade-offs between model size, latency, memory, and accuracy—acting as an authoritative technical counterpart to our research team.

Comfort with AI-assisted development: Coding agents are a normal part of how we work; we expect you to use them well, to give them the context and constraints they need, and to read the resulting code critically and fluently.

Security-first mindset: you think about trust boundaries, secrets, dependencies, and supply chain by default, not as a review step at the end.

Fluent in English with a proven ability to thrive in agile, dynamic teams with fluid, cross-functional responsibilities.

Experience in one or more of the following areas is a plus

Solid experience in Frontend and UX experience

Hands-on experience with model compression: quantization, pruning, distillation, sparsity, or inference-runtime optimization.

Familiarity with quantum computing concepts at a high level, or curiosity about them.

Experience meeting formal security or compliance requirements in a small company.

Experience with distributed training, GPU scheduling, or ML cost optimization.