Currently building AI-powered product management systems

I turn ambiguous problems into products people can use.

Product Manager & Product Owner based in India, building across B2B SaaS, enterprise, telecom and AI — from discovery to release and beyond.

Portrait of Abuzar Ahmed Siddiqui
Abuzar Ahmed SiddiquiProduct Manager · N Force One
LinkedIn ↗
4.5+Years in product
15Person cross-functional team
48Sprint releases shipped
40%Fewer post-release defects
38%Fewer inbound support inquiries
Selected work

Products I’ve built, shipped and shaped.

Each one started with a different kind of ambiguity — and ended with a sharper decision.

Flagship · AI × Product × Engineering

AI Product Management Agent

One intelligent layer across product, engineering, QA and analytics — built to surface what needs attention before it becomes a meeting.

JiraGitHubQA / ExcelPower BIGenAI
pm-agent / sprint-overview
Simulated data
Sprint Health
ON TRACK
AI signal
Strategy × CX × Delivery

Enterprise Telecom Platform

Making a complex enterprise experience more legible for the teams and customers who depend on it every day.

B2BEnterpriseRelease quality
telecom / outcomes
Support inquiries−38%
Post-release defects−40%
Discover→Prioritize→Deliver→Measure

  1. The problem

    Release quality issues and scattered feedback made it hard to see what deserved attention.

  2. Discovery & insight

    Bring together customer friction, support patterns, delivery signals and stakeholder context to find the smallest high-confidence intervention.

  3. The decision & trade-off

    Prioritize clarity in the end-to-end journey over a broad redesign that would add delivery risk without resolving the core customer problem.

  4. Collaboration & delivery

    Product as connective tissue — aligning engineering, QA, architecture, business and customers around an explicit decision and a measurable release plan.

  5. Result

    40% fewer post-release defects and 38% fewer inbound support inquiries.

  6. What I’d do differently

    Build earlier feedback loops around the riskiest assumptions, and keep more qualitative customer context alongside the operational metrics.

Approach

Product is a learning system, not a conveyor belt.

The loop I run every initiative through. Pick a stage to see the questions, artifacts and measures behind it.

Stage 01

Discover

Questions
Artifacts
People
Measure
Product lab

Small bets before big ones.

Prototypes and systems I’m exploring before they become something larger.

● LIVE

AI PM Agent

Product signals stitched together from the tools teams already use.

EXPLORING

Discovery signal map

A lightweight way to find recurring customer friction across channels.

PROTOTYPE

Release confidence

A decision aid connecting scope, risk, QA and customer impact.


Experience

Built through curiosity, context and collaboration.

Ownership, change and lessons — not a list of duties. Education & learning →

Aug 2024 — Present

Product Manager · N Force One

Working across product strategy, discovery, delivery, AI and customer experience to turn complex needs into shippable decisions.

StrategyDiscoveryDeliveryAICustomer experience
Feb 2022 — Jul 2024

Junior Product Manager · InsightX Studio

Built product judgment by partnering closely with cross-functional teams, translating needs into a clear path from problem to release.

RoadmapsAgile deliveryProduct analyticsStakeholders
Writing

Things I’m thinking about.

Practical notes on building products. All notes →

AI Product4 min read

The useful AI agent is a decision surface, not another dashboard.

How an agent earns trust by showing its evidence and uncertainty.

IN DRAFT
B2B SaaS5 min read

What enterprise customers mean when they ask for “more control.”

Behind a familiar request is often a workflow confidence problem.

IN DRAFT
Discovery3 min read

Roadmaps are promises. Learning loops are the product.

Planning systems that leave room for evidence to change the plan.

IN DRAFT
Open to product conversations

Let’s build what’s next.

Have a product challenge, a role, or an idea worth pressure-testing? I’d love to hear about it.