Voxquill is hiring for a team in AI infrastructure. Inference and post-training for open models.

They're past $1B ARR, and some of the fastest-moving AI companies run production workloads on them. You'd own those accounts, from the first POC to production.

Read this part first

• You're based in the US. Remote is fine, or the New York or San Francisco office.

• You've worked directly with customers. Not next to them. You've run the call, scoped the POC and owned the account.

• Your engineering depth is ML or AI, not only backend. Data science counts if you actually built things.

• The process starts with a take-home build, before the conversations. If that's a no, this one isn't yours.

What you'd do

Own accounts from week one. Scope POCs and MVPs with the customer, build them, and get them into production. Deploy, integrate, debug, post-train and chase latency and cost. Run workshops on site with customers. Push PRs back into the platform itself and turn what you hear in the field into product.

Must have

• 5+ years, most of it hands-on with ML or AI systems. Strong Python.

• Production ML or AI systems you built from the ground up and kept running.

• Customer-facing work you can point to with specifics.

• Comfortable with regular in-person customer visits.

Nice to have

• vLLM, SGLang, fine-tuning (SFT, DPO, RFT), Kubernetes, GPU infrastructure.

• Time as an FDE or embedded engineer at an AI-native startup.

• You've started managing, or were about to.

What's on offer

$176K to $228K base, $220K to $285K OTE, plus equity.
OPT and H-1B transfers welcome, TN sponsored

How hiring works

First, you'll have an intro call with us at Voxquill. We'll walk you through the full process before you commit to anything.