Senior Product Manager · AI · Data Platforms · Payments

Building payments, data & AI products used by millions.

I'm Fayazuddin Mohammad — 10+ years building AI-powered data platforms, fintech, and payments products from 0→1 through scale. I ship products that improve decision-making and automate analytics for millions of users across banking and digital payments. Currently at Zeta; previously NPCI, where UPI & BHIM went from launch to 5M+ users in six months. Now looking to lead product in AI, data & payments.

View my work Email me LinkedIn ↗ Resume ↓

Open to product leadership roles · AI / Data / Payments

Fayazuddin Mohammad
Zeta · 4.9 yrs  /  NPCI · 5.6 yrs
Scroll

Impact, settled.

Outcomes across two platforms
0+

Product initiatives shipped across data, payments & AI

0M+

Users onboarded on UPI within 6 months of launch

0%

Faster time-to-insight with AI-powered self-serve analytics

0%

Fewer hallucination incidents on AI features via evaluation frameworks

Where I've shipped.

2016 — Present

Senior Product Manager · Zeta

Nov 2021 — Present

Own product strategy, roadmap, and execution for Zeta's enterprise banking suite — 30+ initiatives shipped across data, analytics, reporting, and governance, cutting client reporting turnaround by 40%.

  • Launched a unified reporting platform — cutting dependency on engineering by 50% and enabling self-serve analytics for 100+ users.
  • Built an AI-powered analytics assistant that answers business questions in natural language — cutting time-to-insight by 60–70% and opening analytics to non-technical users.
  • Defined evaluation frameworks and benchmark datasets gating every AI release — 60% fewer hallucination incidents and zero rollbacks over 12 months.
  • Shipped 50+ data models powering enterprise reporting, regulatory workflows, and customer integrations — accelerating downstream reporting by 40%.
  • Improved platform data reliability from ~85% to 98%+ and cut manual reconciliation effort by 60% through automated data-quality frameworks.
  • Prototyped Data Playground — a self-serve data-discovery hub bringing data dictionaries, sample data, and lineage & relationships for data models into one place.
  • Led product lines across Payments & Disputes, Collections, Customer Analytics, and Financial Reporting — 20+ high-impact solutions tied to business KPIs.
SQLDBTAirflowApache SupersetAzure OpenAIJiraFigma

Senior Associate · NPCI

Apr 2016 — Oct 2021

Owned end-to-end product strategy and rollout for UPI and BHIM — the payment rails India runs on today.

  • Scaled UPI & BHIM from launch to 5M+ onboarded users within 6 months through phased releases and partner integrations.
  • Designed and launched UPI Mandate and the Central Mapper module — establishing the recurring-payments capability now used across the ecosystem.
  • Shipped a rewards & automated dispute-resolution product, growing processed dispute volume 30% while cutting resolution time.
  • Led discovery and alignment across 10+ bank, fintech, and TSP partners, authoring BRDs, PRDs, and TSDs that became the reference spec for integrations.
  • Drove compliance & risk strategy across UPI modules — resolving systemic issues like duplicate debits and limit breaches, cutting post-launch incidents by 40%.
  • Established QA and requirement-traceability frameworks, improving release quality across high-volume transaction systems.
UPI SwitchBHIMMandatesCentral MapperBRD / PRD / TSDQA Frameworks

Selected work.

Three deep dives · AI, Data & Payments
AI ProductZetaIn production

Teaching AI the data before the SQL.

Problem

Every data question queued behind engineering — and early LLM prototypes made it worse: hallucinated tables, broken joins, answers nobody could trust enough to ship.

Decision

Context before generation. The assistant retrieves the right models from DBT metadata — schema, relationships, business definitions — before writing a line of SQL. Golden datasets and an evaluation framework gate every release.

Outcome

Business users self-serve answers in plain English, and accuracy became a managed metric instead of a hope. The pattern now underpins the platform's AI roadmap.

0%

fewer hallucination incidents

0

production rollbacks in 12 months

~0%

of routine data requests now self-served

"AI is only as good as the context you provide — not the model you use."

Data PlatformB2B BankingZeta

One credit data mart, designed for consumption.

Problem

Banks consumed credit-card data through fragmented models with duplicated logic and definitions that disagreed — slow onboarding, engineering on the critical path of every report.

Decision

A layered modelling strategy — raw, consolidated, business-ready — with the consolidated layer as the primary consumption surface. Models designed around business use cases, not source systems, governed by data contracts and documentation standards.

Outcome

Reporting sped up, definitions converged, onboarding stopped waiting on engineering — and the mart became the foundation the AI querying layer was built on.

0%

faster downstream reporting

0+

data models documented & governed

wks→days

customer onboarding time

"Design data for how users consume it, not how systems generate it."

PaymentsNPCIEcosystem scale

Recurring payments for a national rail.

Problem

UPI had no standard way to authorise recurring payments — every PSP improvised one-off flows, and there was no reliable way to resolve users across apps.

Decision

Treat the ecosystem as the user. Designed the UPI Mandate lifecycle — create, modify, pause, revoke — and the Central Mapper spec, then aligned 10+ banks, fintechs, and TSPs on reference BRDs, PRDs, and TSDs that became the integration standard.

Outcome

Recurring payments became a rail-level capability used across the UPI ecosystem, shipped alongside compliance work that cut systemic incidents like duplicate debits.

0+

ecosystem partners aligned on one spec

0%

growth in processed dispute volume

0%

fewer post-launch incidents

"At ecosystem scale, the spec is the product."

What I bring to the table.

Domain depth & toolkit
Payments & CardsUPI · BHIM · Credit Cards

Six years at NPCI building UPI & BHIM at the switch level — mandates, disputes, recurring payments — plus credit-card data products at Zeta.

Data Platforms & AnalyticsMarts · ETL · Governance

Own Zeta's enterprise data platform — 30+ initiatives shipped across marts, extracts, governance, lineage, and self-service analytics.

AI & LLM ProductsNL-to-SQL · Evals

Took NL-to-SQL from concept to production — evaluation frameworks and golden datasets that cut hallucination incidents 60%, zero rollbacks in 12 months.

Regulatory & RiskCompliance · QA

Drove compliance & risk across UPI modules at national scale — resolving duplicate debits and limit breaches, cutting post-launch incidents 40%.

Toolkit

Product

Vision, Strategy & RoadmappingPRDs & SpecsDiscovery & ResearchOKRs & North-star MetricsProduct Analytics & ExperimentationGo-to-MarketStakeholder AlignmentScrum

Data & AI

Data ModellingData GovernanceData ContractsMetadata ManagementRAGAI AgentsPrompt DesignA/B Testing

Tools

SQLREST APIsDBTAirflowApache SupersetFigmaJiraGit

AI-native workflow

ClaudeChatGPTCursorGitHub CopilotAzure AI Foundry

The lab.

AI products & prototypes
In production

Enum

Natural-language-to-SQL assistant with metadata retrieval, query explanation, and conversation memory — analytics for people who don't write SQL.

POC

DataWeaver

Agentic analytics assistant — orchestrated metadata, SQL, execution, and insight agents that turn enterprise reporting into a conversation.

POC

Data Playground

Self-serve discovery hub bringing data dictionaries, sample data, lineage, and model relationships into one place.

Concept

AI Documentation Assistant

Auto-generates schema documentation, business glossaries, and data dictionaries straight from metadata.

Concept

AI Data Quality Assistant

Watches for missing data, freshness issues, schema drift, and anomalies — quality as a continuously monitored product.

How I think.

Principles & operating rhythm

Principles

01

Start with the problem, not the solution.

02

Adoption doesn't fail on missing features — it fails on fragmented experiences.

03

AI should augment expertise — and it's only as good as its context.

04

Metadata and documentation are product features, not support work.

05

Great PMs don't just ship features — they prevent unnecessary ones.

How I work

Customer problem

Where it hurts, and why it matters to the business

Discovery

Workshops with users, SMEs, and engineering

Prioritisation

Impact vs effort — and saying no with evidence

Solution design

Specs, architecture, edge cases, trade-offs

Build

Tight loops with engineering and QA

Launch

Phased rollouts behind evaluation gates

Adoption

Docs, onboarding, enablement

Iterate

Feedback, metrics, and the next bet

Credentials, verified.

Education & certifications

Education

PG Certification, Blockchain TechnologyIIIT-B, Bangalore
PG Diploma in ManagementITM Business School
B.Tech, Computer ScienceJNTU, Kakinada
LanguagesEnglish · Telugu · Hindi · Urdu

Certifications

Professional Scrum Product OwnerScrum.org
Certified Data Science ProfessionalIBM
Intl. Payments System ProfessionalMVLCO
Project Management Certificationi2p2m
Data AnalyticsITM Business School

Recognition, earned.

Awards & growth

Awards

2026
2025

Outstanding Award × 2Highest honour

Zeta's top individual recognition for the year — won back-to-back in 2025 and 2026.

Zeta
2024

Mountain Mover Award

For data-mart segregation, critical reporting, and rapid data turnaround that kept a SOC 1 audit on track.

Zeta
2024

Trailblazer Award

For stepping beyond the day job — independently building Data Playground and an AI-powered query generator.

Zeta
2022

Shining Star Award

Zeta
5 / 5

Top appraisal rating, three consecutive cycles

Rated 5 — the highest — in three back-to-back appraisal cycles.

NPCI

Growth

Zeta · 2021 — Present

Associate Product Manager→ Product Manager→ Senior Product Manager

NPCI · 2016 — 2021

Associate→ Senior Associate

In their words.

What colleagues say

While managing the day to day work seamlessly, Fayaz consistently steps out of his comfort zone, exploring new tools and independently building models that add significant value. The latest one being Data Playground and AI powered query generator.

Parvathi Bhat & Vinay · Zeta · Trailblazer Award citation

Fayaz has delivered outstanding performance this quarter… his support during the SOC1 audit by swiftly aggregating the data requested by the auditors. His proactive approach has made a significant impact in several areas, and his prompt responses have kept everything on track.

Parvati · Zeta · Mountain Mover Award citation