Founding AI Engineer

Founding AI Engineer

New York City (Union Square). In-person. $200,000 to $250,000 plus equity & token budget.

Voiceops is the intelligence layer for consumer-facing businesses. We turn millions of customer conversations into a live model of how the business runs, and use it to power agents that take on the work across sales, product, marketing, and operations.

$12M raised, seed stage company. Customers include Motorola, Kin Insurance, and Capella University.

Voice data is messy, unstructured, and massive. A single client of ours can generate millions of calls per year across dozens of product lines, geographies, and agent populations. Extracting reliable structure from that data, structure accurate enough to run production workflows on, requires multi-level agent systems that discover structure in the data, validate extractions, identify statistical patterns, trigger downstream actions, and continually learn, all autonomously.

What you'd build:

• Architecture. Multi-layer agent systems that autonomously discover structure in raw conversation data, then orchestrate downstream agents to act on it. This is systems design at a level that no other company in our space is operating at.

• Frontier. AI is moving so fast. Our non-technical users are building their own tools with Claude Code! The architecture has to evolve at this pace. You will help us keep the system at the edge of what's possible, beyond what our clients can dream of: new inference strategies, new model capabilities, new ways to get 10x more out of the same team and system.

• Scale. We process millions of calls. The pipeline has to be fast, reliable, and cost-efficient. You'll work on the infrastructure that makes this work.

There is a wide gap between a demo of an AI product and an AI product that actually works for real customers. We are hiring engineers who can build on the right side of that gap.

We increasingly believe the data layer is the whole game in AI. Most tools start with a point solution and try to inject intelligence into it. We are doing the opposite: build the intelligence layer first (from data), and let the point solutions (sales coaching, lead scoring, business insights, CRM updates, agentic workflows, voice agents) fall out as different activations of the same underlying intelligence.

The market demand is here. Our biggest growth bottleneck is engineering velocity. Customers pull us into new features every week, stretch the product by using it for more things (and break our system in the process!), and bring us into CTO and CIO conversations about broader AI transformation. Even our non-technical users are building their own tools on top of us with Claude Code.

Product velocity is where our recent capital is going.

We are building a small, elite team that ships fast and sits in the same room. You will work next to the CTO and CEO, talk to customers directly, and ship in hours and days rather than weeks and months.

You are familiar with how modern AI products are built and shipped. You have worked with LLMs, agents, tool use, and the standard moving parts, and you have opinions about what works and what does not.

You have shipped real software to real users. You can describe in detail something you built, what broke, and how you fixed it. You move across the stack and have product instincts. You do not need a clean spec to start.

If you bring deep experience in any of the following, tell us. We are hiring across a few different shapes of this role, and these are the areas where we want strong owners.

• Building AI agents that operate over large or messy datasets, with thoughtful sampling, summarization, and bounded queries.

• Designing multi-agent systems and pipelines. Layered agents, generator and critic loops, planners and executors, and orchestration where one agent's output drives another's work.

• Agentic chat that reasons over real customer data and can act on it.

• Building AI products that ask very little of the user. This means the UI itself, plus the scaffolding behind it: smart defaults, agent harnesses, and prompt structures that let the agent do the heavy lifting on…