The problem FVM-AI-Deploy solves

Generate -> Compile -> Deploy -> Verify -> IPFS -> Explore

FVM-AI-Deploy was built for the purpose of improving the developer experience on the FVM Chains, For FVM to be used and adopted widely the developer experience has to improve and FVM-AI-Deploy was built with the same goal. Not only does it improve the deployment, interaction and verification process but it also allows generate the contract code automatically based on user request using the power of AI. If you want to deploy a contract asap or are you a beginer on writing contract, FVM-AI-Deploy is a place for you to genearte contract and deploy in a min's.

Writing flaw less contracts are time consuming and also if begginer wanna explore more kind of contracts based on differnet logics, currently no platform that help for that. So with the ultimate power of AI, FVM-AI-Deploy helps to solve the problem by generating contracts based on yours input.

1. FVM-AI-Deploy will generate and deploy the contracts in a mins based on user requirments.

2. It have explore option to explore the depolyed contracts code , abi and bytecode. so that devs can easily export those into there frontend to interact with them.

3. It push Contract code, abi, bytecode to the IPFS using Web3.storage. Explore option interact with the IPFS to fetch these deatils to our platform.

4. Registery Contract on FVM HyperSpace -> 0xA5A51315b449C7026164111ED142E87cd1C865B7. Once contract deployed it register on our registery contract with the details like who owns this contract and ipfs links.

AI mechanism - Current open source AI models are not much good on generating compileable contracts so, FVM-AI-Deploy use fine tune version of open-AI Generate model. Here we tweaked that model more into generate storage related contract development. so our results comes well for storage related contracts. Currently we are working on improve the model more into generation on erc20,erc721 and erc1155 development.

Challenges I ran into

1. Most of the open sourced AI models dont produce compile friendly contracts, comes with many flaws so fine tuning the AI model for some particular purpose was very challenging

2. Current model not good on generating the contracts that more into nfts or complex defi contracts. Improving the model that keeps learn over those contracts.

3. As a data analyst, not good at front-end development. so its takes some time to crack some basic structure.