Campus Labs
2
AM
Arjun Mehta
ISE · 3rd Year

Welcome back, Arjun!

Ready to continue your learning journey? You have 2 active sessions.

22
Labs Available
47h
Total Lab Hours
12
Saved Projects
8
Achievements

Quick Launch View all labs →

AI/ML Training
TensorFlow, PyTorch
Quick
LLM Lab
Ollama, LangChain
GPU
Cloud & DevOps
Docker, Kubernetes
Quick
Cybersecurity
Kali, CTF Platform
Quick

Active Sessions View all →

AI/ML Training Lab
Running for 1h 23m · 4 CPU, 8GB RAM
Running
Cloud & DevOps Lab
Running for 45m · 4 CPU, 8GB RAM
Running

Resource Status

CPU Usage72%
Memory78%
GPU Usage92%
Storage45%

Storage Usage

6.2 GB
of 10 GB · across all labs
Recent files model_v3.pt · 2.1 GB
notebook_lab2.ipynb · 14 MB
Campus Labs
2
AM
Arjun Mehta
ISE · 3rd Year

My Sessions

All your active labs and recent session history

2
Active Sessions
47h
This Month
12
Saved Projects
28
Total Sessions

Active Sessions

AI/ML Training Lab
Running 1h 23m · 4 CPU · 8GB RAM · CIFAR-10 + MNIST loaded
Running
Cloud & DevOps Lab
Running 45m · 4 CPU · 8GB RAM · K8s namespace student-arjun-7
Running

Recent Session History Last 30 days

Campus Labs
2
AM
Arjun Mehta
ISE · 3rd Year

Resources

Industry certifications, saved projects, and recommended learning

Industry Certifications I'm Pursuing

NVIDIA Deep Learning Institute
Generative AI & LLMs track
60% complete9 / 15 modules
AWS Certified ML - Specialty
Eligible after 2 more lab hours
85% prep completeEligible
HuggingFace LLM Course
Completed Mar 14, 2026
CompletedCertificate ready
CompTIA Security+
Cybersecurity foundations
35% complete5 / 14 modules

Saved Projects (12)

Project
Lab
Size
Last edited

Recommended Learning

VIDEO TUTORIAL · 47 min
PyTorch in 90 Minutes
From basics to training a CNN on real datasets. Hands-on, no slides.
DOCUMENTATION
Kubernetes the Hard Way
Bootstrap a K8s cluster from scratch. Same exercises every CKAD candidate runs.
COURSE · 6 weeks
AWS for Students
Free tier credits + a guided path through Lambda, S3, and EC2. Resume-ready.
Campus Labs
2
AM
Arjun Mehta
ISE · 3rd Year

All Labs

Choose from 22 specialised technology labs

AI & Data Intelligence 8 Labs

Emerging Technology 6 Labs

Security, Networks & Enterprise 8 Labs

Campus Labs
Need help?
AM

Select Resource Profile
Session Duration
ℹ You can extend later. Auto-saves every 5 min.
Pre-load Datasets (Optional)

Ready to launch your environment?

Standard profile · 2 hour session · MNIST + CIFAR-10 pre-loaded

Estimated startup time: under 8 seconds

AI/ML Training Lab
1:47:32 left
NVIDIA A30 MIG
train_model.ipynb ×
explore_data.ipynb
Python 3 (PyTorch) | Idle ●
FILE BROWSER
datasets/
mnist/
cifar10/
imagenet/
models/
cnn_v1.pt
resnet50.pt
notebooks/
train_model.ipynb
explore_data.ipynb
utils.py
requirements.txt
My Storage
3.2 GB / 10 GB used
GPU active
CUDA 12.1 · 6.0 GB VRAM
[1]:
import torch # Verify GPU is available via NVIDIA MIG print(f"GPU available: {torch.cuda.is_available()}")
Out:
GPU available: True
[2]:
print(f"Device: {torch.cuda.get_device_name(0)}") print(f"Memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.1f} GB") print(f"PyTorch: {torch.__version__} · CUDA: {torch.version.cuda}")
Out:
[3]:
# Ready to train. Load dataset and define model... # model = SimpleCNN().to(device) # optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
Full AI environment. NVIDIA GPU active. Ready to train.
Foundation Models Lab (LLM)
1:52:14 left
NVIDIA A30 MIG
chain-builder.ipynb ×
rag-pipeline.ipynb
Python 3 (LangChain) | Idle ●
FILE BROWSER
models/
llama-3.1-8b/
mistral-7b/
phi-3-mini/
prompts/
system.txt
examples.json
chains/
chain-builder.ipynb
rag-pipeline.ipynb
vector-db/
embeddings.faiss
My Storage
8.4 GB / 10 GB used
GPU active
CUDA 12.1 · 6.0 GB VRAM
[1]:
# Build a question-answering chain with a local LLM (no data leaves campus) from langchain_community.llms import Ollama from langchain.prompts import PromptTemplate llm = Ollama(model="llama3.1:8b", temperature=0.2)
[2]:
prompt = PromptTemplate.from_template( "You are a teaching assistant. Explain {topic} to a 3rd-year ISE student in 3 bullets." ) chain = prompt | llm print(chain.invoke({"topic": "backpropagation in neural networks"}))
Out:
[3]:
# Next: load PDF lecture notes into a vector DB and build a RAG chain over them...
Llama 3.1 8B loaded · VRAM: 5.4/6 GB · Tokens/sec: 42 · LangChain 0.2.16 · Ollama 0.3.6
All models run on-campus. Student prompts and responses never leave the GPU server.
Cybersecurity Ops Lab
ISOLATED NETWORK
2:45:08 left
Kali Terminal
Wireshark
Target Network
CTF Challenges
KALI TOOLBOX
Reconnaissance
▸ nmap
▸ recon-ng
▸ theHarvester
Exploitation
▸ metasploit
▸ sqlmap
▸ hydra
Wireless
▸ aircrack-ng
▸ kismet
Forensics
▸ autopsy
▸ volatility
Web
▸ burp suite
▸ owasp zap
Lab Network
10.99.0.0/24 · isolated
Targets: 4 vulnerable VMs
No internet egress · safe to attack
student@campus-kali-pod-7:~/ctf-week3$
CTF Week 3 — In progress... Rank 7 / 124 · Streak: 3
Network is fully sandboxed. Pre-built vulnerable VMs. Safe to attack. Zero risk to campus systems.
Cloud & DevOps Lab
ns: student-arjun-7
3:12:44 left
Code Editor
Terminal
K8s Console
GitLab CI
EXPLORER
▾ student-app/
▾ k8s/
deployment.yaml
service.yaml
ingress.yaml
configmap.yaml
▾ docker/
Dockerfile
docker-compose.yml
▾ terraform/
main.tf
▾ .gitlab/
ci.yml
▾ src/
app.py
Cluster Connection
campus-k8s-prod
Context: arjun@campus-k8s
Namespace: student-arjun-7
deployment.yaml ●
YAML · Spaces: 2 · UTF-8
1apiVersion: apps/v1 2kind: Deployment 3metadata: 4 name: student-app 5 labels: 6 app: portfolio-site 7spec: 8 replicas: 3 9 selector: 10 matchLabels: 11 app: portfolio-site 12 template: 13 metadata: 14 labels: 15 app: portfolio-site 16 spec: 17 containers: 18 - name: app 19 image: student-app:v1.2
student@campus-devops-pod-3:~/student-app$ click "Build & Deploy" to run...
Each student gets a real Kubernetes namespace with quota limits. Build, deploy, debug — exactly as in industry.
Lab
2:00:00 left
Lab environment is ready.
Provisioning your environment
Allocating resources...
LIVE DEMO