San Francisco | $200–350K base + up to 1% equity | In-person

The shape of the role

Two modes, not one.

Architecture. Platform primitives, backend services, and data infrastructure built to be reused across every customer.

Customer-facing product engineering. Something in a customer's hands in days, then hardened into core product.

Someone who only does one half isn't a fit.

You're employee #9 at a seed-stage company on six-figure multi-year contracts with public companies, working alongside the cofounder/CTO with real influence on the roadmap. The scope is bigger than it would be almost anywhere else.

About Minoa

<aside> ℹ️

The question every enterprise buyer asks before signing a deal isn't "what does this cost?", it's "what is this actually worth to us, and can you prove it?"

There's no infrastructure for this yet. It's done by consultants at $500/hr, by sales teams guessing in spreadsheets, or it doesn't get done at all. We're building value intelligence: the infrastructure enterprises use to prove what a product is worth. Value agents that uncover what each customer uniquely gains from a product, generate defensible business cases in minutes, and measure ROI across the full customer lifecycle.

Our customers include Snowflake (3,000+ reps on the platform), Ironclad, Vanta, Benchling, Cognite, and Pendo. Zero churn. Team of 8, growing to around 12 this year. We're not optimizing an existing category, we're creating one, and there's no incumbent to displace.

</aside>

The technical challenge

Value intelligence is harder than it looks.

A business case for a $3M Snowflake deal requires stitching together data from Salesforce, the customer's data warehouse, their product telemetry, their financial model, and public-market comps, then reasoning about counterfactuals (what happens if they don't buy?) in a way a CFO will actually sign off on. Every enterprise has a unique value framework, unique data shape, and unique buying committee. The system has to be flexible enough to adapt to each and rigorous enough that a procurement team can audit it.

The hard problems we're actively working on:

• Agent orchestration over messy enterprise data. Getting a value agent to reliably navigate a customer's Salesforce schema, their product analytics, their custom objects, and their financial model, and produce a number a finance team will trust.
• Generalizing customer-specific logic into platform primitives. Every customer wants their own value framework. We have to build the abstractions that make the tenth deployment fast without gluing on custom code that rots.
• Production AI systems that public companies run deals on. Latency, reliability, auditability, and cost all matter because sales leaders are running live pipeline on top of us.
• Turning prototypes into product at speed. The right prototype in front of a customer on Friday often becomes a platform feature next month. You'll be the person who makes that transition work.