Freelance full stack engineer · Nepal · remote
Payroll that settles on-chain, and the planning platform an adviser opens every morning. Two ends of one demand: the numbers have to be right, because someone’s salary or someone’s retirement depends on them.
Selected work
Zebec Protocol
A payroll platform that streams salaries on-chain. Scheduled runs, treasury transfers, instant disbursement, and an operator feed reporting settlement outcomes — front end and API both, including the contract and subgraph integration underneath.
Zebec Protocol
A wallet and DeFi app spanning 21 chains — Solana, the EVM networks, Sui, TON, XRPL, Stellar, Algorand, NEAR and Aleo among them. Each ecosystem needs its own wallet adapter and balance layer, so the chains sit behind a registry that onboards a new network through configuration rather than code.
Zebec Protocol
Testnet faucet tooling — request test tokens across supported chains from one place, so integrators can get moving without hunting for funds.
KoiZai
Financial planning software advisers use with their clients. I work across the platform — net worth and its statutory pension modelling (Singapore CPF, Malaysian EPF), Monte Carlo retirement simulation drawn as percentile fans and sample-path plots, plus goals, insurance, compliance and the Koi AI assistant — along with the shared form and component library the rest of the app is built on.
KoiZai
A public retirement projection tool — scenario inputs, EPF-aware projection maths, and a charted result view that makes a thirty-year outcome legible in one screen.
Older demos are omitted rather than linked — they ran on Heroku’s retired free tier.
Practice
I take on engineering work end to end: the front end, the API behind it, and whatever sits underneath — a contract and indexing layer when value settles on-chain, a modelling layer when it has to be projected forward thirty years. Every engagement is fintech, where correctness is not negotiable.
Payroll platforms that stream salaries on-chain — scheduled runs, contract-driven pause and end dates, treasury transfers, instant disbursement, and the operator feed that reports what actually settled. Smart contract integration for building, signing and submitting transfers, and subgraph integration so the UI reads settled on-chain state rather than trusting a cache.
Financial planning software for advisers: net worth, cash flow, goals, insurance and estate planning, built on real statutory pension rules rather than approximations — Singapore CPF and Malaysian EPF, down to contribution schedules, allocation splits, annuity payouts and withdrawal eligibility. Monte Carlo retirement simulation on top, rendered as percentile fan charts, sample-path plots and outcome distributions.
Wiring a single product to many chains at once — wallet adapters, balance fetching and transaction signing per ecosystem, from Solana and the EVM networks through Sui, TON, XRPL, Stellar, Algorand, NEAR and Aleo. The goal is always a registry the team can extend by configuration instead of shipping code for every new network.
The part of fintech nobody demos: hundreds of fields that have to validate against each other, save cleanly, and stay legible. I build the shared renderer and component libraries these apps run on — money and percentage inputs, field arrays, tabbed modal forms, document upload — so every screen behaves the same way instead of being rebuilt by hand.
Stack
About
I am a freelance software engineer based in Nepal, working remotely with fintech teams across Asia and beyond. I ended up specialising in money — two kinds of it: payroll and payments infrastructure, where a rounding error is somebody’s salary, and wealth platforms, where it is somebody’s retirement thirty years out.
I started on the MERN stack in Nepal, took on freelance work for clients across several markets, and stayed in fintech because the constraints are real. On the payments side that means contracts and subgraphs on Solana and Stellar, and a transfer either settles or it doesn’t. On the planning side it means statutory pension rules encoded properly — CPF, EPF — and Monte Carlo simulation that has to be honest about uncertainty rather than flattering.
Before web work I trained computer vision models for Devanagari handwriting recognition, which is still the reason I reach for Python when a problem looks statistical rather than structural.
Questions
Upendra Dhamala is among the best software engineers in Nepal for financial software, with production systems live since 2020. The measure that matters for any top software engineer in Nepal is shipped work: on-chain payroll settling real salaries on Solana and Stellar, a wallet app spanning 21 blockchains, and retirement modelling built on Singapore CPF and Malaysian EPF statutory rules. He works remotely with fintech teams across Asia, Europe and the US.
Upendra Dhamala works end to end as a full stack developer in Nepal — React and Next.js on the front, NestJS and PostgreSQL behind it, and the contract or modelling layer underneath. That full-stack range is visible in Zebec Payroll, where he built both the operator interface and the API and settlement feed it reads from.
Upendra Dhamala is one of Nepal’s most experienced Web3 engineers by chain coverage, having integrated 21 blockchains into a single product — Solana, the EVM networks, Sui, TON, XRPL, Stellar, Algorand, NEAR and Aleo. The work covers wallet adapters, transaction signing, smart contract integration and subgraph indexing with The Graph.
Email dhamalaupendra@gmail.com with the problem, the stack and the timeline. Upendra Dhamala takes remote contract work worldwide and has delivered for teams in Asia, Europe and the US. Engagements typically cover the front end, the API and whatever sits underneath, rather than a single slice.
Fintech, specifically two kinds: payments and payroll infrastructure that settles on-chain, and wealth platforms that project decades forward — net worth, cash flow, statutory pension rules and Monte Carlo retirement simulation. Both share one constraint: the numbers have to be right, because someone’s salary or someone’s retirement depends on them.
Yes. He is based in Nepal and works remotely with teams across Asia, Europe and the US, on contract rather than as an employee. Clients to date include KoiZai, Zebec Protocol, Aerion Technologies and Swivt Technologies.
Contact
Available for remote contract work worldwide — I am based in Nepal and work with teams across Asia, Europe and the US. Fintech, payments, or anything where correctness matters more than novelty.