Hi, I'm Esha Sharma 👋
֎ AI Product Manager
📊 Turning user problems into product decisions

Curious about AI products, product strategy, and how user trust actually works? You're in the right place.

I specialize in writing PRDs, running product teardowns, and researching how AI products fail their users, then designing the solutions.

What I do
📋 PRD Writing → End-to-end product specs with user research, RICE prioritization, and rollout plans
🔍 Product Teardowns → Identifying systemic failures connected to real business metrics
🤖 AI Trust Research → Documenting how and where AI products fail, with evidence
📊 Data & Analytics → SQL, Python, A/B testing, cohort modeling, retention analysis

Currently working on
AI PM portfolio projects, PRDs, teardowns, and trust research
LLM fine-tuning and AI product behavior analysis
Metrics frameworks for AI-specific product failures
Building in public before being hired

Featured projects
🧠 AI Trust & Failure Analysis — Independent Research
Tested ChatGPT, Claude, and Gemini across 5 real task types
Defined 5-type failure taxonomy with real observed evidence
Built Trust Recovery Framework and 5-metric KPI system including Silent Failure Rate
Delivered VP-level executive memo on confidence signaling as highest-ROI retention lever
https://eshasharma.notion.site/AI-Trust-Failure-Analysis-36599496b51c80f29222d6678d9a7734?source=copy_link

📋 Notion AI — Workspace Memory Layer PRD
Designed end-to-end concept PRD for a missing Notion AI feature
User research, RICE prioritization, AI privacy framework, 6-week MVP scope
https://eshasharma.notion.site/The-Missing-Layer-A-PRD-for-Persistent-Context-in-Notion-AI-36399496b51c80aebc92cb95e78b2980?source=copy_link

📉 Perplexity AI — Product Teardown
Identified 3 systemic product problems connected to retention and trust
Designed 3 prioritized feature recommendations with target metrics
Defined Northstar metric: Weekly Research Return Rate
https://eshasharma.notion.site/Perplexity-AI-Independent-Product-Teardown-35e99496b51c80c18b61db2650082d4f?source=copy_link

📊 Olist E-Commerce Funnel & Retention Analysis
Analyzed 100K+ real-world e-commerce orders using SQL and Python
Identified funnel drop-offs and cohort retention trends revealing ~30–35% month-2 churn
Built Tableau dashboard and delivered 3 prioritized fixes projecting 15–20% retention improvement
https://github.com/promptedbyesha/Olist-Ecommerce-Analysis

Tools and skills
Product
PRD Writing, RICE Prioritization, User Research, Product Teardowns, Metrics Design

AI & Analytics
SQL, Python (pandas, scipy), Tableau, Power BI, LLM Fine-tuning

Business skills
Stakeholder Communication, Decision Memos, KPI Frameworks, A/B Testing

🤝🏻 Connect with me:
📧 Email: workwitheesha@gmail.com
LinkedIn: https://www.linkedin.com/in/workwitheesha/
Portfolio: https://workwitheesha.framer.website/

🤝 Let's Connect!

Got an AI product problem that needs sharper thinking? A team building something in the AI space?
Or just want to geek out about trust metrics and failure taxonomies at 2am? I'm in.

---

⚡ Fun Fact

I can look at an AI product failure and immediately think of three product solutions. 🧠
I've probably written more PRDs than I've taken vacations, that's a metric I'm actively fixing. 📋 → 🏖️

---

Made with 💛 by Esha · turning AI behavior into product decisions, one framework at a time