Payments PM
0%
Loading
HS
Open to PM roles·Delhi NCR · Bangalore

Hello! I'm

Himani
Sharma

Payments & Fintech PM

Consumer-fintech roots at Paytm, product ownership at a payments aggregator, now building with AI. Moving deeper into payments infrastructure.

Ex-PaytmFintech & paymentsCSPO®
Acceptance
UPI · cards
Lending
Postpaid · loans
Collections
& payouts
Settlement
& recon
Payments
UPI
Cards
Settlement
Recon
Claude
MCP
Himani Sharma
Ex-Paytm · Fintech & payments

Who I am

About me

I'm a payments and fintech product manager. At Paytm I worked across UPI (P2M and P2P), lending and co-branded cards. Most recently, as a Product Associate at SabPaisa (a payments aggregator), I shipped from spec to launch, took new products to market, and contributed to the annual operating plan.

I also build with AI, so the products I design, I can ship myself.

01 · What I do

What I do

A payments & fintech PM working across UPI, lending and onboarding. I use AI to build the products I design.

Payments & Fintech

Consumer & B2B

01

B2B product work at a payments aggregator. Consumer-fintech exposure at Paytm: UPI, cards, lending, collections and payouts.

UPI & PaymentsLending & CardsCollections & Payouts

A fintech PM who builds

Comfortable in the codebase

02

I use AI tools to turn product ideas into working software I can test and improve.

Claude CodeCursorMCPRAG + Evalsv0Remotion

Commercial & Growth

Research, GTM & pricing

03

Authored the market research behind the SabPaisa 3.0 products launch: ICPs and personas, competitive analysis, onboarding benchmarking, market sizing, pricing and lead-gen. Worked with product and business leaders to finalize the launch set and align positioning, and turned the research into per-product GTM plans, one-pagers and landing-page reviews. Built team-distributable plugins and programmatic launch videos for internal and external communications.

Market ResearchGTM StrategyPricingPositioning

Reconciliation & Settlement

Self-built proof project

04

PayOps is a project I built to work through reconciliation and settlement: it matches orders, gateway exports and bank settlements, calculates net settlement after fees and GST, and flags 7 types of exceptions. The math is rule-based; AI only drafts the investigation notes, never the numbers.

ReconciliationNet SettlementExceptions

02 · Experience

Career & experience

Product Associate

SabPaisa

2025 – now

Worked across product delivery, go-to-market and partner-bank pitches for the 3.0 payments launch (a new Payment Gateway, the QwikPaisa no-code payments suite, merchant onboarding, settlement, orchestration, POS and SoundBox). Closed a merchant-onboarding gap: built the platform's first re-verification (Re-KYC) flow, frontend and backend, live to about 135 merchants in the opening cycle. Solved for both sides: a risk-based trigger for merchants and a verifier-and-approval flow for internal ops. The call: make the trigger configurable for ops (due-date, last KYC, periodic) rather than hardcode one rule, so re-verification policy could change without an engineering cycle. Was in the core group for the company's AI-first shift and 10-day hackathon, and contributed to the annual operating plan.

In the press: SabPaisa's AI-first launch

Product Associate (Freelance)

YourGold

2025

Integrated Cashfree's UPI gateway for live gold buy/sell and payout flows, building the onboarding, verification and error-recovery logic for reliable disbursements.

Product Operations Associate

Paytm

2021 – 24

Worked across 10+ products at Paytm, leading optimization of Postpaid, Personal Loans and co-branded Credit Cards, and playing a key role in the rollout of UPI across all transaction types (P2M online and offline, and P2P). Drove data-driven product enhancements from user feedback and transaction data, lifting productivity ~93% and quality ~96%. Launched and ran Paytm's product social channels on LinkedIn and Instagram. R&R award, 2022–23.

03 · Selected work

My work

01

PayOps Copilot

Payments Ops · Reconciliation

Payment-ops teams reconcile across three sources that never quite agree, with no clear owner or audit trail. I built a workspace that matches them, computes net settlement after fees and GST, and flags 7 exception types. The key call: keep the math rule-based and let AI only draft the investigation notes, never the numbers.

Tools & features

ReconciliationSettlementStructured OutputPostgres
View on GitHub
github.com/himanisharrma/payops-copilot
PayOps Copilot product screenshot

04 · Toolkit

My tool stack

The tools I use: models, coding agents, agent/RAG infrastructure, and the product + engineering tools around them.

AI & build

Claude CodeAnthropic APILangGraphv0Make.comRemotionElevenLabs / Whisper

Code & IDEs

CursorWindsurf / DevinVS CodeVercelGitHub CLI

Product & analytics

JiraConfluenceNotionFigmaMiro / FigJamMixpanelSQLGrafana

Engineering

ReactNext.jsTypeScriptNodePythonFastAPI / FlaskPostgreSQLAWS (Rekognition)

How I build with AI

Working knowledge

RAGMCPPrompt EngineeringEvalsVector Search & EmbeddingsMulti-agent OrchestrationAgentic WorkflowsComputer VisionGuardrails