The technology stack of a startup AI company
We are a startup company focused on developing a product that leverages AI to create slides. In this document, I will outline our technology stack selection process.
As a technical manager with over 15 years of experience, I've managed to maintain my coding skills throughout the years. Despite my role in a product R&D department overseeing a team of over 50 individuals, I've remained actively involved in the coding process. This is largely due to the supportive technical culture of the company I recently joined, which has allowed me to continue writing high-quality code.
My journey with Rust began in 2020. While mastering the language presented considerable challenges, I persevered through three attempts and eventually became proficient. I introduced Rust to our company and successfully integrated it into our production environment.
Three months ago, I left my previous company to embark on a new product development venture. For our technology stack, I chose Rust, Go, and Vue.js to build our application.
We employed Go for backend development, Vue.js for the frontend, and Rust for WebAssembly to create a document engine. Rust was especially valuable in document format parsing and conversion due to its exceptional performance, which is critical for our needs.
Regarding our database, we opted for MongoDB and Redis for efficient data storage. MongoDB, in particular, is a product I have a strong affinity for.
Currently, a significant part of our AI functionality relies on chatGPT3.5/chatGPT4 combined with SD (Supervised Learning for Document Generation). In our latest release, we introduced multi-agent supervised document content generation, which has yielded impressive results.
This marks my first contribution to the community, and I'd like to summarize our technology stack:
- Backend: Rust + Go
- Frontend: Vue.js
- Database: MongoDB + Redis
- AI Integration: chatGPT3.5/chatGPT4 + SD
If you are also working on an AI application, I encourage you to consider our technology stack, as it offers a compelling balance between research and development efficiency and product quality.
Of course, I'm also interested in learning about the technology stacks your team uses. Each team has its unique style and preferences, so why not share your tech stack with us?
I use chatGPT3.5 to help me revise my grammar.
