The AI built on your codebase
Arsenal turns your past work into an instantly searchable, queryable memory for code.
Live Demo
How do I implement authentication in my firebase project?
Set up a listener that tracks the user's login state and update your app accordingly. You did this in a previous project...
const [user, setUser] = useState(null);
useEffect(() => {
const unsubscribe = auth.onAuthStateChanged(
user => setUser(user)
);
return () => unsubscribe();
}, []);The useEffect sets up a listener using auth.onAuthStateChanged that fires whenever the user's auth state changes. Inside that listener, setUser(user) updates the local state, so your app always knows who's logged in. Returning unsubscribe() ensures the listener is properly cleaned up when the component unmounts.
Project-Aware
Every learning is tied to its Git project, preserving the full context of when and where you learned it.
Smart Search
Ask questions in plain English and get explanations of your past code, not just exact matches.
Auto Sync
Arsenal automatically syncs your learnings on every push, building your knowledge base as you code.
A code editor, made for you
See how Arsenal helps you remember and reuse your best code
Log new code you might use later
Trying out a basic regression in PyTorch? Arsenal saves how you structured your model and training steps.
X = torch.tensor([[1.0], [2.0], [3.0]])
y = torch.tensor([[2.0], [4.0], [6.0]])
model = nn.Linear(1, 1)
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
loss_fn = nn.MSELoss()
for _ in range(100):
pred = model(X)
loss = loss_fn(pred, y)
loss.backward()
optimizer.step()
optimizer.zero_grad()How did I run that simple linear model in PyTorch again?
You logged this basic training loop for a single-feature regression problem:
model = nn.Linear(1, 1)
loss_fn = nn.MSELoss()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
for _ in range(100):
pred = model(X)
loss = loss_fn(pred, y)
loss.backward()
optimizer.step()
optimizer.zero_grad()↳ Loop that fits a simple y = 2x line using gradient descent
Arsenal retrieves and explains your previous training loop so you can reuse the setup without re-Googling it.
Get responses based on YOUR CODE
Arsenal doesn't just find your old code — it breaks it down so you can reuse it quickly and confidently.
Stop losing time to forgotten code.
Get Arsenal and turn your past work into your fastest asset.
Join Now