Job Description

Key Responsibilities:

• Own a focused portfolio, not a sprawling one. You'll carry a capped, ranked set of active initiatives (target: 2-3 max concurrently) with clear launch dates — depth over breadth.
• Build, don't just spec. Prototype AI features directly — LLM-powered workflows, agents, automations — using tools like Python, LLM APIs, RAG pipelines, and no-code/low-code platforms — so initiatives can move without sitting in the dev backlog.
• Lock scope before build starts. Get sign-off on a PRD before development begins; any new requirement after that goes through a lightweight change-request, not a silent scope add.
• Run a real decision cadence. Every initiative has a named decision-owner and a standing checkpoint — no initiative sits in limbo waiting on "someone to decide."
• Partner with Engineering as a peer, not a queue. Work directly with engineering leads on feasibility and sequencing up front, instead of discovering capacity constraints after commitments are made.
• Own outcomes end-to-end — from opportunity sizing through UAT to post-launch metrics — and report actual turnaround time against target, not just status.

Job Requirements

 Qualifications: 

• A critical thinking problem solver who has the technical tools to build, validate and ship
• 3–5 years of product experience, including hands-on work shipping AI-powered features (not just managing a roadmap for them).
• Practical experience with LLM APIs (OpenAI, Anthropic, Gemini or similar), prompt engineering, and retrieval-augmented generation (RAG). You know how to manage AI instead of letting it think for you. 
• Comfortable scripting in Python and querying in SQL well enough to build and test a working prototype yourself.
• A track record of shipping — not just specifying — automation or AI workflows, using tools like n8n, Retool, Zapier, CoWork
• Strong, direct communicator who can hold a scope line and drive a decision to closure, not let it drift.

Skills