Senior Product Manager 0→1 AI, payments & B2B SaaS · country P&L and team
I build products and the businesses around them.
Seven-plus years, five-plus in product, building 0-to-1 B2B SaaS, B2C and AI products across India and Africa. Today I'm Country Head and Product Manager for FuelBuddy in Zimbabwe, running the P&L while keeping release ownership of both platforms.
Results from FuelBuddy, largest scope first. Each started in the field, not in a document.
Market entry, P&L
Led the evaluation of Zimbabwe, Zambia and Nigeria, then launched B2B bulk fuel in Zimbabwe: ~15 suppliers signed in month one, including the national oil company on USD settlement, and 50+ enterprise and regional clients won from a 400+ network. Supplier volumes are allocated across the depots from a daily demand forecast, with zero stock-outs. Case study
B2B platform, 0→1
Validation and compliance rules built from 300+ pump-owner interviews took activation to under half a day. The platform now serves 150+ vendors. Case study
Payments, 0→1
Specified and shipped UPI Intent and QR checkout, piloted with FuelBuddy's own teams in Gurugram, Bengaluru, Hyderabad and Odisha, so failure handling and reconciliation were proven before customers relied on them. Case study
IoT, industrial SaaS, 0→1
Defined 100+ requirements for one platform across tank gauges, dispensers, flow meters, RFID nozzle tags and tank sensors. Measured against the site's own baseline, the six-month pilot cut loss, theft and spillage by about 20 paise per litre, on roughly 2 million litres of diesel a month. Case study
Operations, experimentation
Root-caused failures in IoT-server order sync, route optimisation and slot booking, tested the fixes against the live delivery baseline, and kept the ones that moved the average: delivery time came down from 60 to 55 minutes, and by up to 10 minutes on some trucks.
Commercial terms stay private. Figures are rounded. The LegalTech MVP and Agile process work from 2021–22 is in the career path.
Built on my own time, each from a field problem I've seen first-hand: product docs first, then working code you can open.
5 in 1areas of depot work in one live view
B2B operations SaaS (energy)
Depot operators juggle stock, pricing and fleet fuel logs in spreadsheets. One live view across depots, deliveries, fleet, inventory and finance, with margin always visible.
10languages in one local-first GST ledger
B2B SMB, vertical SaaS, India
Indian textile traders run billing, stock, job work and receivables across notebooks and Excel. A local-first GST ledger with ARA, a built-in AI assistant, in ten languages.
100%of 100 test trips completed, vs 86% for a rules baseline, with 0 price-limit breaches
B2B, AI agents, travel
Tour operators chase hotels and vendors by hand. A multi-agent system where LLM agents read trip requests and vendor replies, then decide to accept, counter or escalate by calling tools, while code enforces the price limits and a person approves every message. Python, LangGraph, tool calling and RAG, measured on 100 synthetic trips against a rules baseline.
Range beyond my day-job domain: a two-sided EdTech trust problem, multi-tenant B2B credit control, and a consumer subscription offer.
B2C EdTech, live
Can exam prep for UPSC, SSC and banking aspirants win a family's trust? Daily targets, mock-test log, guardian sharing and consent flows, in English, Hindi and Hinglish. Live, free, with voluntary UPI support.
B2B distribution SaaS, multi-tenant
Can one queue replace the phone for an FMCG distributor? Per-tenant ordering, credit control, stock and SKU catalogue. The MVP is used in distributor pitches.
B2C subscription
Will people pay for training that adapts set by set? Every logged set adjusts the next one, with pricing pages built to test the offer.
Personal builds, separate from employer work. Pravesh is live; the other demos run on sample data.
Written the way I run product reviews: problem, evidence, decision, result, and what I'd do differently. Commercial terms stay private.
Choosing a market, launching B2B bulk fuel in Zimbabwe, and finding what the product has to be before it has to be software.
The company wanted to expand into Africa, with no local presence, suppliers, banking or customers in any of the candidate countries.
Evaluated Zimbabwe, Zambia and Nigeria for the CEO, then picked Zimbabwe. On the ground, 40+ discovery interviews with transporters, vendors, fuel stations and corporates shaped what to sell first and what the product would need later.
The business before the product: banking, procurement, credit, dispatch and reporting under written SOPs, with ~15 suppliers signed in month one, including global commodity traders and the national oil company on USD settlement. The interviews became FMS and VMS requirements fed through a live feedback loop.
A live B2B business with 8 active depots and 50+ enterprise and regional clients won from a 400+ network, zero stock-outs against a daily demand forecast, and the Country Head role added to the product one.
I scoped an on-site fuel management system for a large enterprise client, hardware and digital payments included, before confirming the hardware could be sourced in-country. It's paused. In a new market, check the supply chain for the product before you check the appetite for it.
A B2B vendor platform for petrol-pump owners at a fuel-delivery startup in India, built from what the pump owners showed us.
Bringing a pump on as a vendor took a full working day of document collection and checks, then a second review the next day, so no new vendor could take orders the day they signed up.
300+ conversations with petrol-pump owners, at their pumps rather than over a survey. They showed which checks repeated for every vendor and could be standardised.
Standard validation and compliance rules inside the onboarding flow, so a submission was checked once, against one rulebook, at the moment of entry instead of the next morning.
Activation came down from over a day to under half a day, and the platform now serves 150+ vendors.
The fix was a rule set more than a feature, and it came from the pumps, not from a document. Every platform I've built since started the same way.
UPI Intent and QR checkout for the B2B platform, piloted internally with FuelBuddy's own teams before wider use.
The B2B platform needed digital payments. The riskiest component was the checkout itself: UPI Intent and QR flows, failure handling and reconciliation against orders. Shipping it untested would have meant finding the failure cases in front of customers.
The B2B platform already had daily users inside the company who paid for fuel on it. Every unknown in the checkout spec could be tested with them at no acquisition cost: intent hand-offs, failed-payment states and reconciliation against orders.
Specify the checkout for the B2B platform and pilot it internally first, with FuelBuddy's own teams in Gurugram, Bengaluru, Hyderabad and smaller clusters in Odisha, so the payment stack was proven before wider rollout.
2,000+ UPI transactions in month one of the internal pilot, and the failure and reconciliation paths got real traffic before any customer relied on them.
Partnered with the GM and CEO on Project Arjun, an EV and oil & gas retail infrastructure plan built on 100+ fleet and EV operator interviews; plan approved by leadership.
Tank gauges, dispensers, flow meters and RFID on one platform, piloted at a major Indian steel producer.
At an industrial site, diesel leaves the tank through loss, theft and spillage that nobody can see, so nobody can price it. The plant knew its monthly consumption; it did not know how much of it was waste.
Site walks at the plant, with the people who dispense fuel as well as the people who buy it. The gaps were at the nozzle and the tank, so the product had to live there: automatic tank gauges, dispensers, flow meters, RFID nozzle tags and tank sensors, 100+ requirements in all.
A six-month pilot measured against the site's own baseline, with the saving as the metric. The buyer could verify the saving at the meter, which is what makes a hardware-linked product sellable.
Savings did not arrive in month one. As the controls and site behaviour settled, the six-month pilot cut loss, theft and spillage by about 20 paise per litre against the site's own baseline, a run-rate of ~₹4 lakh a month by months 5–6 on roughly 2 million litres of diesel a month.
Set the expectation of a ramp in the contract. A saving that climbs over six months reads as a miss in month two unless the buyer was shown the curve first.
PRDs and BRDs written to the same standard, from industrial, fleet and fintech problems.
Field-evidence problem framing, outcome-based pricing, MoSCoW scope, metrics with baselines.
Platform bet, what has to be true, ruthless cuts, gated roadmap.
Turning a logistics service into a repeatable, measurable product.
Two-sided fintech UX, trust and safety, edge-case thinking.
Earlier work, rewritten to the same standard: LegalTech compliance hub PRD and InvoiceDiscount for SMEs.
Today I run a country P&L and an eight-person team. Before that I was one of four PMs working with the CEO, CTO and MD and a 15-engineer team.
Each move added a layer: understanding users, building for them, then owning the economics around the product.
Asked by the CEO to run the Zimbabwe business alongside the product role; reports to the Business Head in India. Sets local release priorities for the fuel management and vendor platforms, writes the requirements taking both into Zambia and Nigeria, owns the P&L, including customer credit and collections, and leads an eight-person team. Coordinates with teams across the UAE, India, Zambia and Mozambique.
Moved to Harare for market entry, business development and FMS product fit: 40+ discovery interviews fed the backlog while I launched B2B bulk fuel from zero, with banking, procurement, credit and dispatch under written SOPs, and improved unit margins through pricing and supplier mix.
One of four PMs working with the CEO, CTO and MD and a 15-engineer team. EV-charging and fuel-retail research (Project Arjun), then the vendor platform, IoT FMS, UPI checkout and delivery speed; also secured the company's first Indian Oil (IOCL) retail station lease and led the Africa market evaluation.
Web and mobile MVP from Figma wireframes to launch within 8 months, the company's first Agile product process (time-to-market down 25%), and A/B-tested onboarding flows while website visitors grew from about 120 to 1,000+.
Pollution-control policy research for the Delhi Government.
Archaeology is fieldwork: you learn to trust what people do over what they say. That habit is why my products start at the depot, the pump and the plant, not in a survey form.
Four questions about the role. For each answer you get the matching evidence from my work, marked as direct or adjacent, linked to where you can see it, and a summary you can send to the hiring manager.
No sign-up and no contact details asked. Your answers are not saved, and the summary only leaves this page if you choose to send or copy it.
The fit check needs JavaScript. Download the resume instead.
Open to Senior PM, Lead PM and Product Lead roles. Home base New Delhi, open to relocation, hybrid and remote, and available to join immediately.
