Cheng En Hsieh ｜ Product Manager ｜ London I ship AI products that move real numbers.

Six years across AI/ML, B2B SaaS and e-commerce in APAC and the UK. Recommendation engines, LLM workflows and B2B2C platforms, owned end to end from discovery to GTM.

View selected work ↓ Get in touch
01

About

What I do, and how I decide what not to build.

I'm a Product Manager based in London. Six years across AI/ML, B2B SaaS and e-commerce, three of them owning an ML recommendation product running in production for 1,000+ merchants, then eighteen months taking a B2B2C SaaS platform from problem definition to a live MVP as the only PM on it. Most recently I led growth and AI automation for a UK platform with 127k+ monthly active users.

What I actually do is find the constraint. At Rosetta.ai the interesting question was never whether the model was accurate. It was which client segment personalisation would pay off for, before we spent a year rolling it out to everyone. At JUKSY the growth ceiling wasn't traffic, it was how many articles an editor could physically write in a day.

I work hands-on with data, model evaluation and agentic tooling (Claude Code, Codex) to de-risk decisions before engineering time gets committed, and I ship my own products on the side to keep that honest. Most AI products fail in the translation to commercial outcomes, not in the modelling. That translation is the part I do.

  • 01Find the constraint before writing the spec

    Every roadmap I've owned started by naming the one thing actually capping the business, whether that was throughput, margin or activation, rather than working down the longest feature request list. It's what stops a team shipping the wrong thing beautifully.

  • 02De-risk before engineering time is committed

    I'll build the rough version myself: a data pull, a model evaluation, a working prototype. The team then commits to a direction that has already been tested, not to a hypothesis in a document.

  • 03Ship to a number, not to a date

    Every launch I've run had a metric attached before it shipped and an honest read on it after: conversion, AOV, output per editor-day, engagement time. If nothing moved, that is also a finding, and it changes the next decision.

02

Selected Work

Five case studies across AI, growth, SaaS and e-commerce. Tap any project to open the full case.

01AI AutomationGrowthLondon ｜ 2026

Banyan Digital — AI Automation & Growth

Owned the growth roadmap for a UK platform with 127k+ monthly active users, then removed the thing actually holding it back by shipping five AI automation workflows that doubled output per editor-day without adding a single hire.

127k+
Monthly active users
2×
Output per editor-day
5
AI workflows shipped
9m 33s
Avg engagement time

Users span the UK, the Americas and Asia. Engagement time is the average per active user, sustained after the analytics rebuild.

02AI RecommendationE-commerce2021–2024

Rosetta.ai — AI Recommendation Engine

Owned the product lifecycle for an ML personalisation engine deployed across 1,000+ e-commerce clients, turning generic "bestseller" widgets into real-time, per-user recommendations.

+660%
Revenue increase
+350%
CVR improvement
+250%
AOV growth
−43%
Inference latency

Commercial figures are the average across the early-adopter cohort migrating from zero personalisation, over their first year; the strongest individual accounts reached 250% AOV and 350% CVR growth. Latency is measured against the previous production solution.

03B2B2C SaaSSole PM2024–2025

RoBard — Marketing Automation SaaS

Sole PM owning end-to-end build of a B2B2C SaaS covering press-release automation, influencer marketing and multi-channel publishing. Shipped MVP to production with a cross-functional team of 5+.

+80%
Ops efficiency
MVP→v2
Shipped in <6 months
5 teams
Cross-functional
04AI ｜ Content Automation2024

AI-Powered SEO Content Pipeline

An end-to-end automation that scaled editorial output 7× by combining Google Trends data with a fine-tuned internal ML model trained on JUKSY's brand voice.

7×
Editorial output
−85%
Time-to-draft
60 days
To SEO results

Output measured per editor per day (3 → 20+), with human editorial review retained on every piece before publication.

05Growth ｜ CRMCampaign2024

LINE OA Member Marketing & Influencer Campaign

A full-funnel LINE OA ecosystem covering member segmentation, automated messaging and a branded B2B2C campaign for Taiwan Banai, an illustrator with 4,000+ crowdfunding supporters.

3-track
Lifecycle segmentation
1st-party
Owned-channel data
+

Side Projects

Products I scoped, built and shipped on my own.

06Live productAgentic automation2026

Credit-card rewards search

Ask it where you're about to spend, and it tells you which of 28 cards pays best, net of foreign-exchange fees, caps and registration rules. Built solo, live in production, and kept current by a scheduled agent that re-checks card terms and commits the corrections itself.

28
Cards in the index
Daily
Autonomous data refresh
Solo
Product, build, ops
Where are you spending?Live
NetflixFlightsShopeeCostcoApple
+8.5%E.SUN U Bear
+4.5%SinoPac Dual Currency
+3.0%Cathay CUBE
+2.0%Fubon Costco Card
−0.2%Taishin Richart
Net reward after FX fees, caps & registration28 cards indexed
Live site ↗ Open the tool ↗
07Data productScoping under constraint2026

Travel Radar — Flight price agent

A price agent that watches a route (origin, an optional stop, destination) and alerts only when a fare crosses a threshold worth acting on. The interesting work was product scoping, not scraping: the obvious data sources had just died.

08LLM productUnit economics2026

Daily Brief — Personalised morning briefing bot

A Telegram bot that writes each reader a different morning briefing covering world news, the London job market and a running finance-literacy series, and lets them change their own settings by simply telling it, in plain language.

03

Experience

Six years of product, across the UK and APAC.

Apr 2026 – Jul 2026

Product & Growth Lead, Operations

Banyan Digital Limited ｜ London, UK ｜ Hybrid

Owned the end-to-end growth roadmap for a UK platform with 127k+ monthly active users across the UK, the Americas and Asia. Shipped five AI automation workflows that doubled multi-format output per editor-day at constant headcount, and rebuilt SEO and GA4 measurement around behavioural funnels.

Apr 2024 – Aug 2025

Product Manager

Juksy Digital Media Co., Ltd. ｜ Taipei, Taiwan ｜ Hybrid

Promoted to Product Manager within nine months. Sole PM taking a B2B2C SaaS platform from problem definition to shipped MVP, covering vision, roadmap, PRDs and user flows, and built the AI content pipeline that lifted editorial output from 3 to 20+ articles per editor per day. Ran cross-functional agile delivery with engineering, design, sales and C-suite stakeholders.

Feb 2021 – Apr 2024

Product Specialist & Domain Expert

Rosetta.ai ｜ Taipei, Taiwan ｜ Remote

Owned the product lifecycle for an AI recommendation and personalisation product deployed across 1,000+ e-commerce clients, from the 10,000+ attribute taxonomy the model ranked against, through the A/B testing framework, to production rollout. Partnered with engineering and data science to cut inference response time by 43%.

Oct 2020 – Mar 2023

Project Manager (Freelance)

Self-employed ｜ Taipei, Taiwan

Cross-industry project coordination, client and supplier relationship management, and e-commerce platform operations, alongside video production credits on commercial and feature projects.

Mar 2020 – Aug 2020

Purchasing Manager

MAKALOT Industrial ｜ Taipei, Taiwan

Managed orders, pricing and quality control with suppliers across China, Hong Kong and Southeast Asia; launched 8 items for the Japanese fast-fashion brand GU.

Sep 2015 – Jun 2019

BA Fashion Marketing

Fu Jen Catholic University ｜ Taiwan

Consumer and market analysis, project management and go-to-market strategy. Winner, 46th Graduation Exhibition; led the end-to-end marketing campaign for the graduate degree show.

04

Capabilities

Product judgement, AI fluency and the technical depth to ship with engineering.

Google Analytics Certified Google Ads Search Certified Google Ads Display Certified Mandarin & Taiwanese ｜ native English ｜ fluent

Product Management

  • Product vision & roadmapping
  • Market & commercial sense
  • Data-driven decision-making
  • Discovery & user research
  • PRD & feature specification
  • A/B experimentation & metrics
  • GTM strategy & launch

AI & Technical

  • LLM integration (ChatGPT, Claude)
  • ML product requirements
  • API & integration scoping
  • Data pipeline scoping
  • AI workflow automation & integration
  • Analytics & BI tools
  • Technical SEO

Tools & Leadership

  • Figma & prototyping
  • Notion / Linear / Jira
  • Google Analytics 4
  • CRM & LINE OA
  • Cross-functional leadership
  • Bilingual EN / ZH
  • Stakeholder communication

Open to new roles in London.

Emailchengen.hsieh197@gmail.com ↗ LinkedInin/markhsieh1997 ↗ Phone ｜ London+44 7352 134 838 ↗