Solo-Deving a Full-Stack Solana Project in 1 Week (Prelude)

A year or so ago, after nearly a decade of working exclusively as a dev in the Ethereum space, I decided to add some tools to my repertoire and become a Solana dev as well. The reasons for this were largely economic, as a lot of the attention had seemingly moved there, and I after posting a few articles about my journey across I was headhunted by a team to start building with them using my newly acquired skills.

We got a lot done over the last 12 months, but unfortunately due to the nature of the organisation (NDA-type stuff), I don’t have much publicly to show for it. Luckily for me though, as lead dev building though trials-by-fire, I did still refine the hell out of my Solana knowledge in that time, and am thus pretty confident in my abilities.

So to prove to myself as much as anyone else who sees it that I have actually got the aforementioned skills, I’m setting myself the challenge of solo-developing an entire Solana project in a single week (ideally only 5 days). By the end I’ll have built the entire stack from scratch, deployed the program to Solana mainnet, have a live frontend and some necessary backend, and you will be able to use it. I’ll also publish a short dev log at the end of each day to keep myself honest about the progress.

No AI Used

I will not be using AI for any of this project. I don’t use AI in the course of development anyway, since when you’re working with large sums of money in smart contract development the sort of sloppiness you get from AI generated code is a recipe for very expensive disasters, but felt it necessary to highlight the lack of AI since the less technical may assume all devs use it.

The Project

The idea of this is to roughly approximate the dev experience of getting a contracting gig with a short timeline. As such I have a rough idea of what I want to build, including the overall design, and various components, but am going to let myself solve the project-unique challenges as they occur.

Without giving away the whole thing this early on, I will be building a fast-paced, competitive, fully on-chain and permissionless game with some game-theory-based gameplay, utilising verifiable-randomness for RNG. Players will stake, earn and win an SPL token that I will create on pump.fun.

The rationale for using a token rather than just native SOL is that this is more likely to be the case for a real project, and the added stress of deploying the token and implementing it under time constraints only enhances the challenge.

I will give more info about the actual game idea closer to the release date, but those are the big details.

The Stack

Smart Contract:
Contract will be written in Rust / Anchor.

Unit Tests:
Unit tests will use LiteSVM and will use the Mocha test framework.

Frontend:
The frontend will be Svelte, with Solana Kit for all the crypto stuff. I may or may not include a rant about why Svelte is infinitely better than React for web3 projects in one of my daily dev log, you will just have to wait and see.

The frontend will be hosted on Github, served by Netlify.

Backend:
I will need to build a backend, both to effectively act as a cache of the on-chain data to minimise traffic to the RPC from the frontend, but also to automate some game events.

The backend will be written in Typescript, utilising Solana Kit for the crypto stuff, and PostgresSQL for the database.

The game will still operate fine without the server, it will be fully-on-chain and permissionless, the server will just make things run more smoothly.

Timeline

To get everything done, I’ve set myself the following timeline

Day 1: Domain Registration with Dummy Site and Contract & Unit Testing
Most of this day will be spent doing contract dev and unit testing, but I need to register the domain and set up a dummy site first in order to get ahead of the Phantom wallet white-listing process so the game isn’t flagged as scary and new when it’s done.

Day 2: Contract & Unit Testing
I expect to finish the contract on day 1 and have allowed myself about a day and a half to just do unit tests.

Day 3: Frontend Web3 Integration and Backend Server
Getting the frontend to basically be a fully working version of the final game, but looking like an ugly skeleton. Also having the whole server built and running locally so I can see it working with the web3 frontend.

Day 4: Visuals and Frontend Aesthetics
Literally everything that makes this feel like a game rather than just feel like you’re interacting with a smart contract. This whole day is essentially just design work and web dev. How sexy this ends up being will depend entirely on whether I’m ahead of or behind schedule at this point.

Day 5: Deploy and launch day
Besides deploying and verifying the program on mainnet, getting the server online with SSL, and launching the live site, I’ll also need to create the token on pump.fun and do a final test run using the real-world version of everything before announcing the launch. This could all go smoothly or this could be the most stressful day of the week.

At dawn, we ride.

It will be a hell of a week, and I will probably consume an industrial amount of coffee to get through it, but I think this is a very achievable goal. Having spent the last year working in a team where the collaborative nature lead to various inefficiencies, I am very keen to just smash something out without any obstacles but my own abilities.

Development begins tomorrow. Wish me luck.

Solo-Deving a Full-Stack Solana Project in 1 Week (Prelude) was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.