How modern software really works.

In-depth writing on system design, AI engineering, DevOps and cloud, plus the AI updates worth your time, explained without the hype. By engineers, for engineers.

Read the blog Browse topics

topics covered: 5
to read, no sign-up: Free

notebook/system-design

01System DesignScaling a read-heavy feed
02AI EngineeringTracing a RAG answer
03AI UpdatesWhat changed this week
04DevOpsFrom plan to rollout
05CloudA three-tier network

System Design01 / 05

Scaling a read-heavy feed

read pathcache misscache-aside

AI Engineering02 / 05

Tracing a RAG answer

userWhy does the support bot cite an outdated policy?
toolretrieve(query, k=4)4 chunks
modelThe top chunk predates the update. Re-index on publish, then re-rank.
evalgrounded · 2 sourcespass

context2.4k / 8k tokens

AI Updates03 / 05

What changed this week

ModelsA new model is out. Read the eval, not the launch post.
ResearchLong context or retrieval: when each one wins.
ToolingUpgrading an SDK without breaking your prompts.

What changed, why it matters, what to do next.

DevOps04 / 05

From plan to rollout

$ terraform plan

+ aws_s3_bucket.assets

~ aws_iam_role.deploy

Plan: 1 to add, 1 to change, 0 to destroy.

$ kubectl rollout status deploy/api

deployment "api" successfully rolled out

Cloud05 / 05

A three-tier network

edgeCDNWAF

vpc · 3 availability zones

publicload balancer

privateserviceservicequeue

datapostgresobject store

Margin note

Cache the read path. Let writes invalidate.

Evaluate retrieval before you tune the prompt.

Every release, three questions: what changed, what it costs, what breaks.

Read the plan before you apply it.

Private by default. Expose only the load balancer.

From the EduGeek notebook: one worked entry for each topic we cover.

Topics

What EduGeek covers

Each topic collects the posts that explain it, from first principles to the decisions you make in production.

System Design
How large systems are built: scaling, caching, queues, databases and the trade-offs between them.
First posts in progress
AI Engineering
Building with LLMs: prompting, RAG, agents, evaluation and running models in production.
First posts in progress
AI Updates
The model releases, research and tooling changes that matter, explained without the hype.
First posts in progress
DevOps
Linux, Git, containers, Kubernetes, CI/CD, infrastructure as code and observability.
19 posts
Cloud
AWS and cloud-native services, networking, cost and security, from first principles.
2 posts

Latest from EduGeek

The newest posts and videos across every topic. Read the explanation, watch the walkthrough, then check yourself with an assessment.

End-to-End DevOps Project: Complete Implementation
2025-10-30 · DevOps
Complete DevOps project: microservices application with CI/CD, Kubernetes, monitoring, logging, and production deployment
Terraform Infrastructure as Code: Complete Guide
2025-10-30 · DevOps
Master Terraform: HCL syntax, modules, state management, workspaces, best practices, and production infrastructure patterns
Advanced Terraform: Modules and State Management
2025-10-30 · DevOps
Advanced Terraform: reusable modules, remote state, workspaces, testing with Terratest, and enterprise patterns
Ansible Automation: Complete Configuration Management Guide
2025-10-30 · DevOps
Master Ansible: playbooks, roles, inventory management, Ansible Vault, modules, and production automation patterns
Advanced Ansible: Dynamic Inventory and Tower
2025-10-30 · DevOps
Advanced Ansible: dynamic inventory, Ansible Tower/AWX, custom modules, plugins, and enterprise automation patterns
Prometheus Monitoring: Complete Observability Guide
2025-10-30 · DevOps
Master Prometheus: architecture, PromQL, exporters, alerting, service discovery, and building production monitoring systems

All posts

Read it once. Understand it properly.

Every post is free to read, with no account needed. Start with the newest writing or pick a topic.

Read the blog