Week 2 — Networking & Cloud Building Blocks

In partnership with

Hi Inner Circle, 

Welcome to Week 2.

Last week we got our basics in place — Git, Linux, billing, and the mental models for cloud.

This week, we go deeper into the backbone of the cloud: networking and core services. Think of it as learning the “roads, locks, and maps” of the cloud. Once you understand these, building on top becomes much easier…

Networking Basics

Your cloud journey is nothing without understanding how systems talk. 5 things you should know:

1. IP addresses & DNS → Every machine has an address; DNS is your internet’s phonebook.

2. Subnets & routing → Decide how traffic flows inside your environment.

3. Firewalls & ports → Control who gets in and out.

4. Load Balancers → Distribute traffic across multiple machines.

5. VPCs (Virtual Private Clouds) → Your own isolated network inside the cloud. Why it matters: Every cloud role — DevOps, Security, or Data — needs networking fluency. Without it, debugging becomes guesswork.

Resource:

• GCP Cloud Networking Fundamentals

• AWS Networking Services

• Practice your skills by exploring this AWS VPC Project on GitHub. 

Compute Services

The machines running your workloads. Know these three categories:

• Virtual Machines (VMs): EC2 (AWS), Compute Engine (GCP), Azure VMs.

• Containers: Lightweight, portable, faster to deploy. (Docker, Kubernetes).

• Serverless: Functions that run on demand. (AWS Lambda, Cloud Functions).

Why it matters:

Picking the wrong compute option = wasted cost and poor performance.

Resource:

• Get a deeper look at the trade-offs on YouTube with Containers vs VMs vs Serverless

• Serverless Project Repo: AWS Hello World Lambda | GCP Hello World Function

• Container Project Repo: Docker Hello World | Kubernetes Hello App

Storage & Databases

Think of this as your cloud’s memory.

• Object Storage: S3, GCS, Blob — for files and media.

• Block Storage: Disks attached to VMs.

• File storage: Shared file systems.

• Databases: Relational (Postgres, MySQL) vs NoSQL (MongoDB, DynamoDB).

How to think about the databases and their use cases:

Why it matters:

Storing logs, application data, or backups incorrectly leads to scale nightmares later.

Resource: Watch this video explaining Object, Block, and File storage and this video on SQL vs NoSQL.

Mini Project: Build a photo-sharing app → Images go into S3 (or GCS) and metadata into a SQL database.

• AWS Photo Sharing Workshop

• GCP Node.js Upload to Cloud Storage + Cloud SQL

Identity & Access Management (IAM)

The lock and key of the cloud.

Authentication vs Authorization

Why it matters: Most cloud breaches happen not from fancy hacks, but from weak IAM setups.

Example: giving developers full admin access when they only need read-only - I mean why to create extra trouble:)

Resource:

• Watch this - IAM Deep Dive with practical examples

• Mini Project: Create a role-based access model — Admin, Developer, Auditor. Apply least privilege and test boundaries.

  • AWS IAM Hands-on Policies & Roles

  • Cloud IAM - Hands-on Lab

Monitoring & Logging Basics

You can’t fix what you can’t see. 3 things to focus on:

1. Metrics: CPU, memory, network usage.

2. Logs: Every request and error gets recorded.

3. Alerts: Be notified before users complain.

Tracing While metrics tell you what's happening (e.g., "CPU is at 90%") and logs tell you why (e.g., "Request to database failed"), tracing shows you the end-to-end flow of a single request as it moves through a distributed system. It stitches together a timeline of events, helping you pinpoint latency and errors across multiple microservices.

Why it matters: Cloud infra without monitoring is like flying blind. 

Resource:

• Watch this YouTube video on the difference between logs, metrics, and traces. 

• Mini Project: Deploy a sample app with built-in monitoring and logging.

  • AWS Example: AWS Observability Workshop → Complete hands-on with CloudWatch metrics, logs, X-Ray tracing, and dashboards.

The AI Angle (What’s Different in 2025)

Five years ago, learning networking and IAM was enough. Today, with AI-native workloads everywhere, the same foundations apply in new ways:

• Networking for AI: Training models needs high-bandwidth, low-latency interconnects. Think AWS EFA or Google Cloud’s HPC networking.

• Compute for AI: Beyond CPUs, you’ll need to understand GPUs and TPUs. Picking the right accelerator = huge cost savings.

• Storage for AI: Vector databases and data lakes are now as common as SQL. AI apps rely on embedding stores just like they do on object storage.

• IAM for AI: Service accounts now authenticate AI pipelines. Least privilege matters even more when models interact with multiple datasets.

• Monitoring for AI: You don’t just track CPU — you track GPU utilization, model latency, and drift.

Why it matters: If you want to stand out, don’t just say “I know VPCs and IAM.” Show how you can apply them to AI/ML pipelines.

Action Items for Week 2

• Create your first VPC and subnet in AWS/GCP/Azure/OC.

• Deploy a test app via both containers and serverless.

• Create IAM users with restricted roles.

• Set up a basic monitoring alert on a cloud VM.

Conclusion

That’s your TL;DR for Week 2. Networking, compute, storage, IAM, and monitoring aren’t optional — they are the backbone of every cloud system.

But in 2025, the twist is clear: the same foundations now power AI-native platforms. So this week, don’t just practice the basics.

Ask yourself: how would these choices change if I were running an AI workload? That perspective alone will set you apart from most beginners.

Next week, we’ll jump into automation and Infrastructure as Code — the shift from clicking in the console to managing cloud like a true engineer.

-V

Go from AI overwhelmed to AI savvy professional

AI keeps coming up at work, but you still don't get it?

That's exactly why 1M+ professionals working at Google, Meta, and OpenAI read Superhuman AI daily.

Here's what you get:

• Daily AI news that matters for your career - Filtered from 1000s of sources so you know what affects your industry.

• Step-by-step tutorials you can use immediately - Real prompts and workflows that solve actual business problems.

• New AI tools tested and reviewed - We try everything to deliver tools that drive real results.

• All in just 3 minutes a day

Join 1M+ pros