SignalForge

market forecasting tool

Architecture and technologies

---

Note

I decided to use python since most AI and ML libraries are available easily also it is way faster to develop

Concurrency -> Go

---

Goroutines makes this election almost indisputable.

1. Easy to develop concurrent tasks
2. Escalability
3. Eficiency -> In terms of execution time and memory resources

AI for predictions -> Python

---

Has a widely used variety of libs which makes the developing experience easier. After having a functional product we could migrate this part of the software to C++ if needed since we will have to do lots of operations and we might have to optimize it.

Web scraping -> TBD

---

Options:

Python

Pros:
    + Easy to develop
    + Wide variety of libs, no proxy needed for HTTP requests
Cons:
    - GIL limits the concurrent scraping
    - Its not possible to execute multiple python programs concurrently -> Not eficient

C++

Pros:
    + Lightweight
    + Max optimization
Neutral:
    * The amount of libraries is intermidiate, not as wide as python's
Cons:
    - Difficulty
    - Manual management of proxies for rotational requests

Go

Pros:
    + Full Go stack
    + Easy to debug and maintain
Cons:
    - Limited scraping tools
Neutral:
    * Crafting a scraping library might be a good idea for the Go community