Get EdgeLab running in under 5 minutes.
- Docker Desktop installed and running
- 4GB free RAM
- 2GB free disk space
cd edgelabpython3 test_setup.pyThis checks:
- Docker is installed
- All files are present
- Assignment files are valid
docker compose upFirst run will take 2-3 minutes to:
- Build Docker images
- Pull base images (Python, Java, Alpine)
- Initialize database
- Start services
Once you see:
ui-1 | You can now view your Streamlit app in your browser.
api-1 | * Running on http://0.0.0.0:5000
Open: http://localhost:8501
- Click "Browse Problems"
- Select "Two Sum"
- Write this solution:
def solution(nums, target):
seen = {}
for i, num in enumerate(nums):
diff = target - num
if diff in seen:
return [seen[diff], i]
seen[num] = i
return []- Click "
▶️ Run Tests" - Wait 3-5 seconds
- See results!
SELECT department_id, salary
FROM employees e1
WHERE (
SELECT COUNT(DISTINCT salary)
FROM employees e2
WHERE e2.department_id = e1.department_id
AND e2.salary >= e1.salary
) <= 3
ORDER BY department_id, salary DESCpublic class Solution {
public static boolean isPalindrome(String s) {
s = s.replaceAll("[^a-zA-Z0-9]", "").toLowerCase();
int left = 0, right = s.length() - 1;
while (left < right) {
if (s.charAt(left) != s.charAt(right)) {
return false;
}
left++;
right--;
}
return true;
}
}Press Ctrl+C in the terminal, then:
docker compose downAnother service is using port 5000. Either:
- Stop that service, or
- Change the port in docker-compose.yml
Docker Desktop isn't running. Start it and try again.
The API takes ~10 seconds to start. Wait a bit, then refresh.
First run downloads Docker images. Subsequent runs are much faster.
docker compose down -v
rm database/edgelab.db
docker compose upTo enable AI-powered code feedback:
- Install Ollama: https://ollama.ai
- Pull Phi-3:
ollama pull phi3:mini
- Restart EdgeLab
The system will automatically detect Ollama and use it.
- Read README.md for full documentation
- Check ARCHITECTURE.md to understand the design
- Look at assignment JSON files to see the format
- Try creating your own assignment!
If something doesn't work:
- Check
docker compose logsfor errors - Run
python3 test_setup.pyto verify setup - Make sure Docker has enough resources (Settings → Resources)
- Use Chrome or Firefox (best Streamlit support)
- Keep terminal open to see logs
- First execution per language is slower (image pull)
- Hidden tests never show details - that's intentional!