Senior Solutions Engineer

About the Role

You'll be our first Solutions Engineer. Your job: make customers succeed, from their first training job to a production model.

Our customers are computational scientists, bioinformaticians, and ML engineers working the hardest problems in drug discovery and protein modeling. You connect what they need to what our platform does.

This role is for someone who already understands this science and loves the craft of making a model better. You will spend your days inside training runs and binding data, collaborating with scientists and engineers across biotech, pharma, and academia. The work is deep. The problems are real.

What You Will Do

• Get customers to their first result. Help customers move their datasets onto the platform. Shrink time to first training job with getting-started guides, sample notebooks, and self-serve paths.
• Make stuck models move. This is the heart of the job. A customer's training run is failing, or slow, or stalling on a novel target. You find why. You debug PyTorch and Hugging Face configs, clear pipeline blockages, tune GPU workloads, and guide training code onto new GPU hardware.
• Speak to the data with authority. A customer asks what our antibody-antigen or TCR-pMHC datasets can do for their model. You answer. You know which dataset fits which problem, and why real binding measurements beat public data on hard targets.
• Shape what gets built. Turn the hurdles and requests you see into signal for engineering. Help decide what to build and what to buy. What you learn from customers steers the roadmap.

Qualifications

• You understand this science. Protein, antibody, or immune-repertoire ML. You know what binding, affinity, and specificity mean, and why a model trained on public data fails on novel targets. You can hold your own in a room full of researchers on day one. A degree is not the point. Understanding is.
• You can read and debug ML training code. Real PyTorch and Hugging Face code. Building and fixing the model run itself, not summarizing an error log with an AI tool.
• 5+ years being the technical person people seek out to make hard things clear. Time spent briefing partners, guiding collaborators, fielding the hard questions, or owning the customer conversation. Deep technical skill, and the gift for explaining it to people inside and outside your field.
• Cloud and HPC fundamentals. AWS or GCP, Linux, containers, and hands-on time training models on GPUs. You understand how infrastructure choices shape a training run. That is what lets you help a customer whose job is slow, stuck, or scaling badly.
• Strong communication. Clear written skills for technical documentation, runbooks, and integration guides.
• Startup comfort. You write the first draft of missing documentation and build the initial demo. Ambiguity does not stall you.

Bonus Points (Not Required)

• You have gotten training code running on a new generation of GPU hardware.
• You have worked somewhere nothing was built yet, and you built it.

Location note: this role is based in Bellevue, WA. No relocation assistance available for this role.