Hello, I'm Ganesh Kumar, and I'm building LiveReview — a blast-radius aware AI code review built for your business-critical systems. Star us to help devs discover the project, give it a try, and share your feedback to help improve the product.
Let's first build the simplest possible regression problem ourselves. This will let you see every part of the neural network.
Let create a data for linear regression
We'll create:
y=2x+1
So our training data is:
| x | y |
|---|---|
| 1 | 3 |
| 2 | 5 |
| 3 | 7 |
| 4 | 9 |
| 5 | 11 |
This is intentionally simple because we already know the answer:
y=2x+1
Our neural network's job is to discover:
w approx 2
and:
b approx 1
Linear Regression Function
Our model is simply:
x
│
▼
┌─────────────┐
│ Linear │
│ │
│ y = wx + b │
└──────┬──────┘
│
▼
ŷ
In PyTorch:
import torch
import torch.nn as nn
# Dataset
x = torch.tensor([
[1.0],
[2.0],
[3.0],
[4.0],
[5.0]
])
y = torch.tensor([
[3.0],
[5.0],
[7.0],
[9.0],
[11.0]
])
# Model
model = nn.Linear(1, 1)
print("Weight:", model.weight)
print("Bias:", model.bias)
At this point PyTorch randomly initializes:
w
and:
b
Maybe you get something like:
w=0.37
b=-0.12
So the initial model might be:
ŷ =0.37x-0.12
Obviously, that's bad.
And It will randomly change for every run.
Training Neural Network
loss_function = nn.MSELoss()
optimizer = torch.optim.SGD(
model.parameters(),
lr=0.01
)
for epoch in range(1000):
# Forward pass
prediction = model(x)
# Calculate MSE
loss = loss_function(prediction, y)
# Calculate gradients
optimizer.zero_grad()
loss.backward()
# Update weight and bias
optimizer.step()
if epoch % 100 == 0:
print(
epoch,
loss.item(),
model.weight.item(),
model.bias.item()
)
Eventually you'll get approximately:
w=2
b=1
So the model learned:
ŷ =2x+1
Wrapping Up
Finaly this is how it looks like when combined all the steps together:
Dataset
│
▼
x = [1,2,3,4,5]
y = [3,5,7,9,11]
│
▼
┌───────────────┐
│ Linear │
│ │
│ ŷ = wx + b │
│ │
│ w = ? │
│ b = ? │
└───────┬───────┘
│
▼
Prediction
│
▼
MSE
│
▼
loss.backward()
│
▼
Gradients
│
▼
optimizer.step()
│
▼
Update w, b
│
└──────→ repeat
There is no activation function yet.
That's intentional.
Once this is completely clear, we can introduce a hidden layer.
Then we'll change the dataset to something that cannot be represented by a straight line, such as y=x^2.
Your team's attention is limited, and the deluge of AI-generated code is making it harder to keep production reliable and secure without slowing you down.
I'm building LiveReview, a blast-radius aware AI code review built for your business-critical systems.
Instead of presenting every diff with equal emphasis, LiveReview scores each change by blast radius — how far its impact reaches through your call graph — so you can focus attention where it actually matters.
Spend code review effort where business risk is highest — not spread evenly across every diff.
⭐ Star it on GitHub:
HexmosTech
/
LiveReview
Blast-Radius Aware AI Code Review for Business-Critical Systems
LiveReview: Blast-Radius Aware AI Code Review for Business-Critical Systems
LiveReview is an AI code reviewer that scores every hunk of a diff by blast radius: how far a change reaches through your call graph, how much persistent state it touches, and how well-tested it is. A 3-line change to a shared auth check can outrank a 300-line UI tweak. Your team's attention goes to the highest-risk code first, not spread evenly across every diff.
blast-radius-demo.mp4
LiveReview's Blast Radius & Review Priority scoring, live in the diff viewer.
The exact math, not a black box
Visualize blast radius at a glance
Every factor that feeds the score
Here's the goal:
- A 3-line fix in a function used by 40 other files, that also writes to a database, should score high.
- A 300-line UI change in one file, fully covered by…
Click below to try LiveReview with your codebase:
Top comments (0)