Peach Pilot Product Manager

Atlanta, GA (Buckhead)  |  Hybrid (3 Days In-Office)  |  7+ Years Experience

Most AI companies sell tools. We transform how businesses run.

Peach Pilot builds a platform that ingests everything about how a company operates — every system, every process, every signal — and constructs a Company Brain: a living knowledge graph that connects people, decisions, and outcomes across the entire organization. We deploy 92 pre-built AI agents that work together across every business function, governed by humans at every critical step. The system gets smarter with every interaction.

We don’t sell software licenses. We embed into a client’s operation, learn their business in weeks, show them what’s broken backed by their own data, and redesign their highest-impact business functions with AI. Insurance is our first vertical and our first client engagement is already scoped and funded — with expansion across multiple verticals planned on the same platform and methodology.

Peach Pilot is co-founded by Mario Montag (Predikto acquired by a Fortune 50; McKinsey, PwC) and JP James (Hive Financial Assets, Georgia Tech, TITAN 100). We have a working platform with live infrastructure and a proven data-to-insights methodology.

This is the role that decides what we ship — and why.

As Senior Product Manager, you own the product surface across the Company Brain, the 92-agent library, and the workflows that take a client from kickoff to live deployment. You decide what gets built next, in what order, and for which client problem. You report directly to Mario Montag, CEO and Co-founder, and you operate as a peer to the engineering and QA functions.

This is not a roadmap-curation role. We are early, the platform is live, and the first client engagement is scoped and funded. We need a PM who can land on day one, get into the data, sit across from an insurance executive, and ship product against real customer pain — fast. Visionary roadmaps without immediate customer value are not what this role is about.

You will be the voice that translates insurance domain pain and field signal into a prioritized, defensible product plan — and the person who keeps the team focused on the work that delivers value to the client in front of us, while protecting the platform for every client that comes after.

Own the product roadmap across the Company Brain, agent library, and integration layer. Decide what we build next based on what the active client engagement needs, what the platform needs to scale, and what unlocks the next vertical. Defend trade-offs with data, not opinion.

Go deep on the insurance customer’s actual operating pain — claims, underwriting, distribution, policy servicing, agency operations. Translate that pain into product requirements the engineering team can build into the platform. You are the person who knows the domain cold.

Define what individual agents do, where the human-in-the-loop sits, and how agents coordinate across a workflow. Partner with engineering on agent runtime decisions and with QA on the eval frameworks that prove agents are doing what we say they do.

Run discovery directly with insurance executives and operators — not through a layer of researchers. Validate hypotheses with the client’s own data. Bring back findings that change the roadmap when the data says they should.

Partner with engineering on what’s feasible and what’s next, and with QA on the quality bar before anything reaches a client. This is a startup — lanes will blur, work will cross over, and you will own outcomes that don’t fit cleanly into a job description. The right person is energized by that, not threatened by it.

Our buyers and users are CEOs and operations leaders. Sit across from them, understand their business pressure, and turn that into a product story they can stake their own credibility on. Translate complex platform decisions into language that builds confidence.

AI/LLM: Anthropic Claude, OpenAI GPT, LiteLLM (multi-model routing)

Agent Orchestration: Custom-built agent runtime with reinforcement learning loops, inter-agent coordination protocols, and continuous learning across a 92-agent hierarchy

Backend: Python (FastAPI), async agent runtime

Data & Graph: Memgraph, Neo4j, Qdrant, PostgreSQL, Redis

Integrations: Nango (700+ connectors)

Infrastructure: Google Cloud Platform, Docker

You don’t need to build in this stack — but you need to be fluent enough to make smart product decisions about it, and credible enough that the engineers building it want you in the room.

7+ Years in Product. You have shipped product at companies where the customer feedback loop is short, the stakes are real, and the roadmap moves. You have owned outcomes, not just outputs.

Early-Stage Tested. You have been pre-product-market-fit or pre-Series-B before. You know how to operate when the playbook isn’t written yet, when the customer is right in front of you, and when shipping the right thing this week matters more than a clean two-quarter roadmap.

AI-Fluent. You have hands-on product experience with LLMs, agents, prompt design, and the realities of model output uncertainty. You understand why a Company Brain matters, how a knowledge graph differs from a database, and what “good” looks like in agent behavior. AI experience is the priority for this role.

Insurance Domain (Strongly Preferred). Background in insurance — carrier, broker, MGA, insurtech, agency operations, claims, underwriting, or distribution. You understand how the industry actually runs, where the pain is, and where AI can take real cost or time out of the workflow.

Genuinely Technical. You can read Python, work in a terminal, write SQL, and reason about an API. You don’t ship production code — but engineers respect you because you understand what they’re building and why.

Consulting DNA or Equivalent. Comfortable in an executive room. Skilled at navigating organizational politics, managing stakeholders with competing priorities, and protecting the work from noise.

Bias Toward Shipping. You prefer to put something in front of a customer and learn from their reaction. You are deeply uncomfortable with quarter-long roadmap exercises that produce no shipped product.

Even Better If: You have shipped agent-based or workflow-automation product in production. You have built product on top of knowledge graphs or vector databases. You have led product through a vertical expansion at a platform company.

You are joining a proven founder’s second company with established domain credibility. We have a working platform with live infrastructure and a first client engagement already in motion. You will have access to production data, live workflows, and real compliance requirements from day one.

Every engagement makes the platform smarter. Every client’s data enriches the knowledge base for the next one. You will not be running discovery exercises against a hypothetical user — you will be in the room when AI meets the real business, and what you decide directly shapes what the company becomes.