AI Conversational Solution for the Textile Industry (0-to-1)

Lead AI Product Manager → Head of Product/AI · Frontier.cool Inc. · B2B SaaS Digital Textile Platform

AI Product & Architecture · 2025

Owned product planning and led delivery of a conversational AI product with the engineering team, contributing code from first commit to v1.0 in under six months. The launch delivered 3x faster responses, followed by reliability improvements.

First commit to v1.0
<6 mo
Faster responses at v1.0
3x
Chat timeouts
Eliminated
Major features shipped
6+

The Problem

Textile sourcing involves searching across product information and evaluating material options. The product needed to make that workflow easier through conversation while remaining fast and reliable.

Product Tradeoffs & Decisions

  • Scoped the initial release around a focused sourcing workflow, with response speed and reliability used to assess delivery.
  • Kept capabilities modular so individual parts of the experience could be evaluated and improved.

Technical Approach

  • Combined conversational AI with domain-specific retrieval and a responsive user experience.
  • Improved application performance and deployment reliability through hands-on engineering.

Cross-functional Execution

  • Owned the product roadmap and delivery of 6+ major features.
  • Coordinated product and engineering work from scope through release.
  • Stayed hands-on in Python and API code alongside the engineering team, continuing product evolution after promotion to Head of Product/AI.

Measured Outcomes

  • Shipped v1.0 in under six months from the first commit.
  • Delivered 3x faster responses at v1.0.
  • Improved reliability and eliminated chat timeouts.

Technologies & capabilities

  • LLMs
  • Python
  • Product Strategy
  • Performance Engineering