- Vishakha Sadhwani
- Posts
- 7 Weeks to Job-Ready: A Free, Week-by-Week Tech Roadmap
7 Weeks to Job-Ready: A Free, Week-by-Week Tech Roadmap
Python, DSA, backend, DevOps, cloud, AI, and system design: one video playlist, one practice resource, and one project per week, all 100% free.
Kushal Khandelwal
October 05, 2026
Most roadmaps hand you 200 links and wish you luck. This one gives you seven weeks, a few videos per week, a place to practice, and a project to build.
The projects connect. The API you build in Week 3 gets containerized in Week 4, deployed in Week 5, and gets an AI feature in Week 6. By Week 7 you have one strong portfolio project instead of seven half-finished ones.
How to use this list: Plan on about 20–25 hours a week. Watch the videos at 1.5x and skip what you already know. Spend most of your time on the practice and the project. That's where the learning actually happens.
Week 1: Python, Git & the Command Line
Everything else in this roadmap is built on these three. Get comfortable writing code, saving it with Git, and moving around a terminal.
YouTube videos
The Missing Semester of Your CS Education — MIT — the shell, Git, and scripting lectures. The stuff colleges assume you already know.
Practice
OverTheWire: Bandit — a free game where every level teaches you a Linux command
Exercism — Python track — small exercises with optional free mentor feedback
Build something: A command-line expense tracker that saves data to a JSON file. Follow the roadmap.sh spec, then compare with jmlc643/expense-tracker. Push it to GitHub with a clean README.
Week 2: Data Structures, Algorithms & Big-O
This is what coding interviews test. You don't need to master it in a week, but you need to start.
YouTube videos
Data Structures Easy to Advanced — William Fiset — focus on arrays, hash tables, stacks, queues, and trees
Practice
NeetCode Roadmap — solve 15–20 easy problems (arrays, hashing, two pointers). Every problem has a free video explanation.
Build something: An LRU cache and a small in-memory search engine that indexes text files and ranks results. Reference implementations: TheAlgorithms/Python. More ideas: build-your-own-x.
Week 3: Backend APIs & Databases
Most engineering jobs involve moving data between an API and a database. This week, the main project begins.
YouTube videos
Practice
SQLBolt — interactive SQL lessons in the browser
PostgreSQL Exercises — real queries against a real schema
FastAPI Tutorial — the official docs, and some of the best docs in Python
Build something: A REST API (a bookmark manager or job-application tracker) with PostgreSQL, JWT auth, input validation, and pytest tests. See how real projects are structured in fastapi/full-stack-fastapi-template and zhanymkanov/fastapi-best-practices.
Week 4: Docker & CI/CD
"It works on my machine" stops being an excuse. Package your app so it runs anywhere, and automate your tests.
YouTube videos
Practice
Docker Get Started — the official hands-on guide
Play with Docker — a free Docker playground in your browser
Build something: Containerize your Week 3 API with docker-compose (API + Postgres), then add a GitHub Actions pipeline that runs lint and tests on every push. Reference setups: docker/awesome-compose.
Week 5: Cloud, Deployment & Kubernetes
Get your project off your laptop and onto the internet, and learn how containers run at scale.
YouTube videos
AWS Certified Cloud Practitioner (2026 edition) — freeCodeCamp — watch the core services: EC2, S3, IAM, RDS, VPC
Practice
Killercoda — free Kubernetes and Linux labs in the browser
kind — run a Kubernetes cluster locally, for free
roadmap.sh DevOps — use it to see what's ahead
Build something: Deploy your API on a free tier, then run it on a local Kubernetes cluster with kind. See how a real multi-service app is wired with GoogleCloudPlatform/microservices-demo.
Heads-up: AWS's free tier asks for a card. If you use it, set a billing alert on day one.
Week 6: AI Fundamentals & AI Engineering
Understand what's happening inside a neural network and an LLM, then build something real with one.
YouTube videos
Intro to Large Language Models — Andrej Karpathy — for more depth, follow up with Deep Dive into LLMs
Optional: Building micrograd — Andrej Karpathy
Practice
Build something: Follow Python RAG Tutorial with Local LLMs — pixegami to build a "chat with your PDFs" app. Code: pixegami/rag-tutorial-v2.
Keep it free: use Ollama for both the embeddings and the model. Then wrap it in a POST /ask endpoint inside your Week 3 API.
Week 7: System Design, Interview Prep & Capstone
Learn how systems scale, prepare for interviews, and finish with one end-to-end project that ties everything together.
YouTube videos
ByteByteGo channel — short videos on caching, load balancing, and rate limiting
Practice
System Design Primer — the most-starred system design resource on GitHub
Tech Interview Handbook — resume, behavioral, and coding interview prep
Build something: The End-to-End DevOps + AIOps Project playlist. It covers Docker, Kubernetes, CI/CD, GitOps, cloud infra, monitoring, and logging, so it doubles as a review of Weeks 4–5. Code: vishakhasadhwani/devops-ai-playbook.
Video Quick List
Every video in one place, in case you just want the links:
Week 1 — Python · Git & GitHub · Missing Semester
Week 2 — Big O · Data Structures · DSA
Week 3 — SQL · Python APIs
Week 5 — AWS Cloud Practitioner · Kubernetes
Week 6 — Neural Networks · Intro to LLMs · Deep Dive into LLMs · micrograd
Week 7 — System Design · ByteByteGo
Projects — RAG with Local LLMs · DevOps + AIOps playlist
Be Honest With Yourself
Seven weeks gives you real foundations and a deployable, AI-enabled project. Most people still need a few more months of steady DSA practice, more projects, and applications before they land a role.
If you start now, you'll finish in late November. Use December to repeat Weeks 2 and 7, polish your GitHub, and start applying.
Don't just copy the reference repos. Build your own version first, then compare. That's the part interviewers can tell.
All links are free courses, videos, docs, or open-source repos. Video availability changes over time, so if a link has moved, a quick search on the title will usually find the current version.