Founding AI Engineer

About the role

NemoRouter is a private, early-stage company building an LLM gateway that gives every customer all the enterprise features — guardrails, A/B tests, prompt management, evals, budgets — free for life, with 240+ models behind one API key. It's live, in production, with real customers. This is a founding engineering role, and we're mostly hiring new graduates who want to build real infrastructure from day one. Compensation, honestly: your first three months are expenses-only while we build to revenue, then base salary begins — set by your country and location within a confidential range (we don't publish a number because it's location-dependent). After one year of continuous service you're recognized with co-founder status, and your founding equity matures only after three years of continuous service. We're a private company, so this is private-company stock, not publicly traded. You'll need short personal runway for the first three months; in exchange you get real ownership and a genuine seat at the table.

What you'll do

• Build the routing core — cost/latency routing, provider fallback, and multi-tenant isolation across 240+ models
• Own the safety layer — guardrails, prompt management, A/B testing, and evals that ship to every customer
• Own the credit + billing engine — reserve/settle, per-org and per-key budgets, rate limits; money-correct end to end with no negative balances
• Talk to customers, cut scope, and own outcomes — founding means no fixed lane

Required experience

• New graduates welcome — strong CS fundamentals and a fast learner
• Solid Python and/or TypeScript; comfortable picking up Postgres
• You ship fast and take real ownership, comfortable with ambiguity and no playbook
• Care about correctness where money and multi-tenancy meet
• Short personal runway for the first three months (expenses-only period)