EzDeploy

Demo

About

The problem it solves ### 🚨 **The Problem EzDeploy Solves** The Web3 ecosystem is facing a critical bottleneck in a **global shortage of Solidity developers**, with only ~30,000 active devs as of 2025, while there are more than 27 Million software developers around the world. This scarcity has led to massive inefficiencies **startups spend over $15K and 8+ hours per smart contract deployment** (Upwork data) and **not focusing on development**, often relying on unsafe, **copy-pasted code**, which has contributed to over **$3.5B in hacks** from 2022–2024 (Immunefi). Meanwhile, interest in tokenizing real-world assets (RWAs), like supply chains and real estate has already skyrocketed, **growing 600% in 2023 (BCG)**, but adoption is throttled as 90% of SMEs **lack technical resources to deploy secure**, customized smart contracts. Without democratizing deployment, the industry risks leaving **trillions in value untapped**, with **less than 1% of the $326T RWA** market currently onchain. ### 🛠️ **How EzDeploy Solves It** 1. **🧱 No-Code Smart Contracts** * Deploy **pre-audited OpenZeppelin ERC-20, ERC-1155, ERC-721 and NFT templates** in under **5 minutes**, cutting costs by up to **90%** for first-step development in industries that want to implement blockchain. * Empowers non-devs: e.g., a coffee farmer tokenizes harvests without writing a single line of code. * *Proof*: [Sepolia deployment](https://sepolia.etherscan.io/address/0xF57233D6cD049045551596c27f4bb1cc72Faaf86#code) shows a real ERC-20 contract with tax logic created via EzDeploy. 2. **🤖 AI-Guided Customization (Ohara AI)** * Converts natural language prompts (e.g., *“Make a token with 5% tax”*) into **error-resistant, optimized Solidity**. * Reduces code mistakes by up to **70%**, and supports smart logic (like auto tax exemptions) in beta testing. * Outperforms rigid competitors like Thirdweb by dynamically adapting to user needs. 3. **🏗️ Industries-Focused Templates** * Purpose-built modules for **fractional real estate**, **supply chain tracking**, and other RWA use cases. * Targets the **massive unmet demand** in the RWA space, bridging offchain assets into secure onchain contracts. **For more detail, check our pitch deck:** https://youtu.be/Kuu69UBa0SE

How it's made

Challenges we ran into ### 🧱 Challenges We Ran Into Building **EzDeploy** on top of **Ohara AI** and integrating with **Thirdweb SDK** was far from smooth. At first, we made the mistake of trying to develop everything at once frontend, backend, and AI logic inside Ohara’s prompt system which led to broken apps. The UI would often appear functional, but the backend returned persistent errors like “sandbox not found.” After 30+ failed attempts in just two days, we learned to **prioritize backend logic first** and **scale complexity gradually**. This allowed Ohara’s AI to generate more stable and functional code. We also faced significant challenges translating natural prompts into **secure and valid Solidity logic**. Early versions of Ohara often produced faulty contracts, especially for advanced patterns like deflationary tokens or custom tax logic. To improve accuracy, we **trained Ohara on OpenZeppelin templates** and added **rule-based filters** to block unsafe patterns. This increased our contract compile success rate from \~60% to **over 95%**. Contract integration with the **Thirdweb SDK** posed additional friction. The AI model often relied on outdated syntax and patterns, leading to mismatches with the current SDK version. We initially tried switching models, but that didn’t help. In the end, we had to build a **separate wrapper app** and **manually instruct the AI** to avoid incorrect patterns. This involved multiple rounds of trial and error visible in our prompt history. Gas optimization was another critical pain point. AI-generated contracts initially used inefficient opcode structures that led to expensive minting costs. To address this, we integrated **Ethers.js gas estimators** into the generation loop and introduced a “**Gas-Saver Mode**” that cut mint gas usage by more than half. Finally, **Base chain compatibility** required additional engineering. Ohara didn’t support `.base` name resolution or smart wallet hooks. We overcame this by **forking Ohara’s SDK**, adding Base RPC endpoints, and embedding **OnchainKit** into our deploy pipeline. We also separately deployed contracts using the **Thirdweb CLI**, then linked them with our AI interface. Through persistent iteration, debugging, and teamwork, we transformed a chaotic AI-driven workflow into a powerful no-code deployment platform making smart contract deployment more **accessible, secure, and efficient**.