Web3 development for applications people can actually use
GegoSoft builds the application layer of Web3 — dApps, wallet flows, token-gated products and on-chain marketplaces — from Madurai, India. Our advantage isn't the contracts; plenty of teams can write those. It's that we come from a UI/UX and product background, so the wallet connection, the failed-transaction state and the onboarding actually work for someone who has never signed a transaction before.
Last reviewed: September 2026
- Who we are
- GegoSoft Technologies (OPC) Private Limited, a software company in Madurai, Tamil Nadu, India, established 2017. Eighteen engineers. Web3 is one practice alongside Laravel, React, Next.js and mobile development.
- What Web3 development means here
- Building the applications that sit on top of a blockchain — decentralised apps, wallet integration, token-gated access, NFT and asset marketplaces, on-chain identity — plus the conventional backend, admin tooling and integrations that every real product still needs.
- What we build with
- React and Next.js on the front end, wagmi and viem or ethers.js for chain interaction, Solidity contracts on Ethereum and EVM chains, and Laravel or NestJS for the off-chain services.
- How AI changes the work
- Our team uses Claude Code daily. On Web3 projects it mainly buys deeper test coverage, contract review passes and documentation. It never approves a deployment, and value-holding contracts go to independent audit.
- What we don't do
- Play-to-earn token economies whose returns depend on new players joining, high-yield investment programmes, and launch-token projects with no product behind them. We say no to these regularly.
- Typical timeline
- Six to ten weeks for a working dApp with wallet flows on testnet. Sixteen weeks or more for a marketplace with independent audit, fiat on-ramp and admin tooling.
What is Web3, in plain terms?
Web3 is a way of building applications where the user's account, assets and data live on a public blockchain rather than in the company's database. The practical consequence is that a user can take their identity and their assets to a different application, and the company running the app cannot unilaterally delete or seize them.
That's the whole idea. Everything else — tokens, NFTs, DAOs, wallets — is machinery built to make that one property work.
It's a real architectural choice with real trade-offs, not a replacement for the web. Web3 applications are slower, cost money per write, and are harder to change after launch. In exchange, users get portability and verifiable ownership. Whether that trade is worth making depends entirely on what you're building, which is the first thing we work out with you.
How Web3 differs from a conventional web application
The honest version, including the parts that are worse.
| Aspect | Conventional web app (Web2) | Web3 application |
|---|---|---|
| User account | Email and password in your database | A wallet address the user controls and can use elsewhere |
| Who holds the data | You do, on your servers | Public state on-chain, plus off-chain storage you still run |
| Cost per write | Effectively zero | A gas fee, paid by the user or sponsored by you |
| Speed | Milliseconds | Seconds, and the user must approve each transaction |
| Fixing a bug | Deploy a patch | Contracts may be immutable; needs an upgrade path planned upfront |
| Deleting user data | Straightforward | Effectively impossible on-chain — a privacy law problem |
| What the user gains | Convenience and speed | Portability, verifiable ownership, no reliance on your goodwill |
Most successful products are hybrids: on-chain for ownership and settlement, conventional infrastructure for everything else. That's how we build them.
What we build
The application layer, and the unglamorous systems around it that decide whether a product survives its first month.
Decentralised applications (dApps)
The full product — front end, contract interaction, transaction states, error handling and the off-chain services it depends on. Built as a real application, not a contract with a page attached.
- REACT
- NEXT.JS
- WAGMI
- VIEM
Wallet integration & account abstraction
Connection flows for MetaMask, WalletConnect and embedded wallets. Where onboarding friction is the risk, we use account abstraction so users can start with an email and never see a seed phrase.
- WALLETCONNECT
- ERC-4337
- EMBEDDED WALLETS
NFT & digital asset marketplaces
Minting, listing, bidding, royalties and transfers, with metadata handled properly and an admin interface your team can run without a developer on call.
- ERC-721
- ERC-1155
- IPFS
Token-gated products & memberships
Access, content or features unlocked by holding a token or credential. Useful for communities, licensing and subscription models where the membership should be transferable.
- SIWE
- ERC-20
- SOULBOUND
Smart contracts for the application layer
Escrow, royalties, revenue splits, staking and governance — written to be readable and testable, then prepared for independent audit before anything touches mainnet.
- SOLIDITY
- FOUNDRY
- HARDHAT
- OPENZEPPELIN
Web3 product consulting
A short paid engagement ending in a written recommendation on architecture, chain choice, custody model and regulatory exposure — including "build this without a blockchain" when that's the right call.
- ARCHITECTURE
- CHAIN SELECTION
- COST MODELLING
Most Web3 products fail at the interface, not the contract
The contract is usually the part that works. What breaks adoption is everything around it: a connection modal that offers eight wallets to someone who has none, a transaction that fails silently, gas explained in gwei to a person who wanted to buy a ticket. GegoSoft was founded by a designer, and we've spent nine years building interfaces for school administrators, church volunteers and construction estimators — people with no patience for a bad flow.
- Onboarding without seed phrases — email or social login via account abstraction, with self-custody available for users who want it
- Honest transaction states — pending, confirming, failed and reverted each explained in words, not hashes
- Gas that makes sense — costs shown in local currency, sponsored where the business model allows
- Recovery paths — what a user does when a transaction stalls, a wallet disconnects, or they switch device
- An admin panel that exists — because someone on your team has to moderate, refund and support daily
An AI-assisted team, on code where mistakes are permanent
All eighteen GegoSoft engineers use Claude Code as daily tooling. On a Web3 project that matters more than usual, because a deployed contract can't be patched on a Tuesday afternoon. AI buys us the thoroughness that used to be reserved for well-funded protocol teams.
Test coverage at a level the budget wouldn't normally allow
Unit tests, edge cases and fuzzing harnesses for every contract. Writing the hundredth test case now costs a fraction of what it did, so we write it.
Extra review passes before a human starts
Every contract read for reentrancy, access control, integer handling and oracle dependency ahead of engineer review. A second and third set of eyes, cheaply.
Front-end states nobody remembers to build
Wrong network, rejected signature, insufficient balance, stalled transaction, wallet disconnected mid-flow. Enumerated and implemented rather than discovered in production.
Documentation written alongside the code
NatSpec comments and architecture notes produced as we go. Auditors work faster against documented contracts, and audit hours are the expensive part.
Where the line is: AI writes drafts, tests and analysis. It does not approve merges, decide architecture, or authorise a mainnet deployment. Named engineers own every contract, and anything custodying real value goes to an independent third-party audit before launch.
When Web3 earns its complexity — and when it doesn't
We assess this in the first week of any engagement, and we regularly recommend against it. A week of consulting is cheaper than discovering it after the build.
Web3 is the right choice when
- Users must be able to take assets to a competing application
- Ownership needs to be verifiable by anyone, not just by you
- Value must move between parties without you as intermediary
- Rules should execute automatically with neither side able to intervene
- Your users already hold wallets and expect self-custody
- Your product would be less trusted if you could quietly change the record
Build it conventionally when
- Your users have never used a wallet and don't want to
- Personal data is involved and erasure rights apply
- You need frequent, low-value writes at negligible cost
- The product must be corrected or reversed by support staff
- The token exists mainly to raise money rather than to do a job
- "Web3" is in the brief because an investor asked for it
Web3 terminology, defined without the jargon
If you're evaluating proposals from several agencies, these are the words you'll see and what they actually mean.
| Term | What it means | Why it matters to you |
|---|---|---|
| dApp | An application whose core logic runs in smart contracts on a blockchain rather than on a company's server. | It keeps working even if the company stops hosting the website. |
| Smart contract | Code deployed to a blockchain that executes automatically when its conditions are met. | Removes the need for a trusted middleman — and can't easily be fixed if wrong. |
| Wallet | Software holding the private key that proves ownership of an address. | It's the user's account. Lose the key, lose the account, with no reset link. |
| Gas | The fee paid to have a transaction processed by the network. | A real per-action cost. Someone pays it — decide early whether that's you. |
| Account abstraction | A standard (ERC-4337) letting a smart contract act as a user's account. | Enables email login, social recovery and sponsored gas — the main fix for onboarding. |
| Token gating | Restricting access to content or features based on what a wallet holds. | Memberships that users can transfer or resell without your involvement. |
| DAO | An organisation whose decisions are made by on-chain token-holder votes. | Governance in code. Legally ambiguous in most jurisdictions, including India. |
| Layer 2 | A network built on top of Ethereum that batches transactions to cut cost. | Where most consumer applications belong — Ethereum gas is prohibitive for them. |
How a Web3 project runs with us
Sequenced so the irreversible decisions come after the reversible ones.
Feasibility and custody model
Does this need to be on-chain? Which chain, what does a transaction cost at your expected volume, who holds keys, and what regulatory exposure that creates. Written up, and yours regardless.
Product and interface design
User flows including every failure state, wallet strategy, and whether account abstraction is needed. Designed before contracts, because the interface constrains the contract more than people expect.
Contract architecture and threat model
Contract structure, upgrade strategy, access control and an explicit list of what an attacker would try. Agreed before a line of Solidity is written.
Build on testnet
Contracts and application built together with AI-assisted test generation running alongside. Deployed to testnet early so you can use the product rather than read about it.
Independent audit
An external firm reviews anything holding value. We prepare the codebase, manage the process and remediate findings. Not optional on our projects.
Mainnet, monitoring and support
Deployment with verified source on the explorer, monitoring, alerting and a support arrangement covering the period where real problems surface.
A product company that also builds on-chain
We've shipped and maintained our own software since 2017 — GegoK12 school ERP across twenty-plus schools, ChurchCMS, BOQ Manager, ChemTracker, eTender and Daybook. That experience is why we're useful on Web3 projects: we know what it costs to support a product two years after launch, and we design for that from the start. The chain is one component. We build the rest of the system too.
Looking for blockchain development instead?
If your need is protocol-level — permissioned networks, Hyperledger consortiums, tokenisation infrastructure or smart contract auditing rather than a consumer application — that's the other page.
Blockchain development servicesNot sure you need a blockchain at all?
Plenty of products briefed as Web3 are better built conventionally. We build those too — React, Next.js, Laravel and NestJS — and we'd rather do that well than build the wrong thing.
Web application developmentFrequently asked questions
What is Web3 development?
What's the difference between Web3 development and blockchain development?
How much does Web3 development cost?
Do users need a crypto wallet to use a Web3 application?
Which blockchain should we build on?
Can a smart contract be fixed after it's deployed?
Is Web3 development legal in India?
Do you build play-to-earn games or launch tokens?
Where is GegoSoft located and do you work internationally?
Start with the feasibility question.
Tell us what you're building and who it's for. In a week we'll tell you whether it belongs on-chain, which chain, what it costs to run, and what the interface has to solve.