https://github.com/disk0dancer/hse-se-paps
This repository contains implementation lab works
https://github.com/disk0dancer/hse-se-paps
Last synced: 4 days ago
JSON representation
This repository contains implementation lab works
- Host: GitHub
- URL: https://github.com/disk0dancer/hse-se-paps
- Owner: disk0Dancer
- License: mit
- Created: 2024-11-06T08:34:03.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-18T13:18:33.000Z (about 1 year ago)
- Last Synced: 2025-06-15T22:38:27.893Z (about 1 year ago)
- Language: Python
- Size: 459 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# hse-se-paps
This repository contains implementation lab works
# Tesis
ru: Server-side application for code generation and integration with OpenAI API compatible servers, including development environments and continue.dev plugins
en: Server-side application for code generation and integration with OpenAI API compatible servers, including development environments and continue.dev plugins
# TO DO
- [x] Initial repository setup
- [x] Lab Work 1 : Requirements
- [x] Lab Work 2 : Architecture design
- [x] Lab Work 3 : Methods and Classes level design
- [x] Lab Work 4 : API
- [x] Lab Work 5 : Docker +
- [x] Lab Work 6 : Patterns
# Technical stack
- Python 3.13
- FastAPI
- PostgreSQL - database
- ORM
- JWT authentication
- Alembic - db migrations
- UV - python package manager
- VLLM - for llms inference
> Vllm is used to run LLMS on Apple Silicon, docker image was built from source and published on hub. [DOCKER IMAGE](https://hub.docker.com/repository/docker/disk0dancer/vllm-arm).
>
> To run VLLM inference on another platform update [compose.yaml](/compose.yaml).
# Quick start
### 1. Install dependencies
```bash
make install
```
### 2. Run dev
Use local postgres or container, change connection config at `.env`.
```bash
make dev
```
### 3. Run Docker Compose
```bash
docker compose up -d
```
### Check out [Makefile](/Makefile) for more commands and details.