Technical AI Product Manager

Ruby Labs

1d•Remote

About The Position

At Ruby Labs, we are building Direct-to-Consumer products in the AI category. We’re looking for a Technical AI Product Manager to own and scale the integrations and connector ecosystem that powers our AI features — the third-party APIs, app integrations, and MCP (Model Context Protocol) servers that let our products reach beyond the model itself. This is a high-ownership, technical role. Your core mandate is to scale connectors — growing the number, quality, and reliability of integrations available to our users — while making data-driven decisions about what to build, in what order, and how to measure success. You’ll operate inside an AI engineering squad, working shoulder-to-shoulder with engineers on prompt systems, structured outputs, agentic workflows, and evaluation, and collaborating closely with product, growth, data, and billing teams. We’re looking for someone technical enough to read API docs and talk to engineers without a translator, who treats AI tools as a core part of their daily workflow, and who measures success in outcomes — not effort.

Requirements

4+ years in product management, with a strong track record on technical, platform, API, or integration products.
Demonstrated end-to-end ownership of products — from hypothesis through production and iteration.
Solid technical fluency: comfortable reading API documentation, understanding data schemas and JSON, and partnering with engineers without needing everything translated.
Practical understanding of how LLMs work and what modern AI products involve — prompts, structured outputs, agents / tool use, and their limitations.
Working familiarity with the concepts behind MCP (Model Context Protocol), connectors, or integration / developer-platform products.
Hands-on, daily use of AI tools (Claude or similar) for real work — this is how the team operates; it is not optional.
Strong analytical skills and hands-on proficiency with a product analytics tool (Mixpanel preferred) — funnels, cohorts, and dashboards.
Excellent communication and stakeholder-management skills, with comfort in a remote, asynchronous environment.

Nice To Haves

Direct, hands-on experience with MCP — building, integrating, or shipping MCP servers or clients.
Experience growing an integrations marketplace or connector ecosystem (scaling both breadth and reliability of third-party integrations).
Experience with AI gateways / model routing (OpenRouter or similar) and LLM observability / evaluation tooling (Langfuse, LangSmith).
Familiarity with the TypeScript / Node.js / Next.js ecosystem.
Experience in a startup or fast-paced, high-iteration product environment.

Responsibilities

Own and drive the roadmap for scaling the connector ecosystem, including the number, quality, and reliability of integrations (MCP servers, third-party APIs, and app integrations) that power AI features.
Prioritise integrations and connectors based on user demand, business impact, and engineering effort.
Write clear product specifications and acceptance criteria for new connectors, working closely with AI engineers.
Define and own the framework for evaluating, onboarding, monitoring, and maintaining connectors over time.
Translate AI capabilities, including LLM features, agentic workflows, and tool use, into clear, buildable product requirements.
Partner with the AI engineering team on prompt systems, structured outputs, and evaluation pipelines, ensuring product requirements are reflected in technical design.
Make pragmatic build-versus-buy decisions and define the scope of integration infrastructure.
Own features end-to-end, from discovery and specification through QA, launch, and post-launch iteration.
Define success metrics for connectors and AI features, including adoption, reliability, latency, cost, retention, and engagement impact.
Design and run experiments and A/B tests, making ship, iterate, or kill decisions based on quantitative results.
Build and maintain dashboards in Mixpanel and use observability tools such as Langfuse to monitor AI and connector performance and health.
Surface actionable insights and recommendations to engineering and leadership teams on a regular cadence.
Own and prioritise the integrations product backlog.
Collaborate closely with AI engineering, growth, data, and billing teams to deliver initiatives reliably and on time.
Communicate technical trade-offs, priorities, and roadmap decisions clearly to both technical and non-technical stakeholders.
Use AI tools (Claude and others) as a core part of the daily workflow for prototyping, specification writing, analysis, and problem-solving.