About Cheaper Inference
Cheaper Inference is a drop-in, OpenAI-compatible API that charges developers about 30% below provider list price. We help our customers save money in two ways:
- Marketplace pricing. Providers with spare LLM capacity compete on price. We send each request to the cheapest healthy offer and pass the discount on to customers.
- Fewer wasted tokens. Our optimization pipeline uses current research to cut tokens from each request. Every customer can see exactly what each request saved.
Our supply of inference capacity limits how far we can grow. The more good, reliable, low-cost capacity we have, the bigger the discount, the more models we can offer and the more customers we can serve.
The role
You'll be the first person focused full-time on the supply side of the business. You'll find, sign, onboard and manage our relationships with inference providers. These include GPU clouds, model labs, resellers, enterprises with reserved capacity they don't use and neoclouds. You'll be accountable for how much capacity we have, what it costs, how reliable it is and how many models it covers.
You'll work closely with the founders and engineering. In this job, commercial deals and technical details go together: a good deal on a slow or unreliable endpoint is a bad deal.
What you'll do
- Build the supplier pipeline. Find and rank providers with surplus or discountable capacity across models, regions and hardware. Run outreach from first contact to a live integration.
- Negotiate and close. Set pricing, volume commitments, rate limits, SLAs, payment terms and data-handling terms. Structure deals that keep us below list price while staying fair and sustainable for providers.
- Onboard providers. Work with engineering to connect new providers to our routing layer. Meet latency and throughput targets and report accurate usage.
- Manage live relationships. Track each provider's uptime, latency, error rates, fill rates and how competitive their prices are. Run regular business reviews, handle escalations and expand the relationships that work.
- Protect supply resilience. Avoid relying too heavily on one provider, keep backup capacity for our most popular models and plan ahead for demand spikes and new model launches.
- Track the market. Follow GPU and token pricing, new model releases, provider roadmaps and competitors. Turn what you learn into a supply strategy and pricing recommendations.
- Own trust and compliance with partners. Make sure providers meet our standards for data retention, privacy, security (e.g., SOC 2) and acceptable use. Work with legal on contracts and DPAs.
- Shape the marketplace. Help design incentives, pricing mechanics and provider-facing tools and processes that attract and keep good capacity.
What success looks like
- First 90 days: A mapped list of target providers, at least 3–5 new providers signed and live, and a clear view of the health of current supply.
- First year: A deep, diverse provider base covering all major models, measurable gains in price and reliability, and a repeatable process for bringing on new providers.
What we're looking for
- 4+ years in partnerships, business development, supply or procurement, ideally at an AI infrastructure company, cloud or GPU provider, or a two-sided marketplace.
- An existing network across inference providers, GPU clouds or model labs, or a clear plan for building one fast.
- A strong commercial sense. You're comfortable with unit economics, margin, usage-based pricing and capacity commitments.
- Enough technical depth to discuss tokens per second, time to first token, rate limits, quantization, context windows and API compatibility with engineers on both sides.
- A track record of negotiating and closing complex contracts.
- Comfort with data. You can build the dashboards and models you need to manage supply (SQL or spreadsheets).
- An owner's mindset and comfort with ambiguity at an early-stage company.
Nice to have
- Experience with open-weight model serving (vLLM, TGI, SGLang) or GPU capacity markets.
- A background in marketplace design, trading or commodity procurement.
- Familiarity with enterprise security and compliance reviews.
Why join
- Compensation: $150,000 - $250,000 + equity
- You'll own a function that directly sets our product's price, quality and growth.
- You'll work at the center of a fast-moving market for AI compute.
- You'll join a small, high-leverage team with real ownership.