Write code from your first variable to a small service, learn to work in a shared repository, practice the problems coding interviews ask, then design systems. Every exercise runs and is checked, and the pathway ends in projects, a timed assessment across the tools and the interview.
Python, Git, Coding interview patterns, Python for software engineers, PostgreSQL for engineers, System Design, Coding with AI agents, Building with language models
Units
36 units, 146 lessons
Checked exercises
778
Cross-tool projects
2
Final assessment
10 tasks, 65 minutes
About 50 hours of lessons, from the estimate on each lesson.
Your progress
Nothing done yet. Progress saves in this browser as you finish lessons, and in your account when you sign in.
Take the steps in this order. Each step is a set of units from one course, not the whole course: the units this role uses. Open a unit to start its first lesson.
1
Programming in Python
One language you know well matters more than five you have read about. Python is a good first one: you learn values, collections, control flow and functions, then classes and code that handles bad input without crashing.
Every software team keeps its code in Git. You commit, inspect history, stage only part of a change, branch and merge, and set work in progress aside, all in a terminal that behaves like the real one.
Most software interviews include a timed coding round. These units teach the patterns behind most of those problems: hashing, two pointers, sliding windows, stacks, binary search, linked lists, trees and greedy choices. You run your code against tests each time.
Production code is more than a correct function. You model a domain with types, use the standard library well, put clear boundaries around a service and keep the code easy to change.
Design tables with constraints, choose indexes for a workload and test transaction boundaries in PostgreSQL. Read execution plans and practice safe schema changes before a service depends on the database.
From the mid level up, interviews ask you to design a service. You turn requirements into estimates, choose request paths and storage, define interfaces, and handle shared state and background work on a board that draws what you describe.
Practice product-specific instruction files and a bounded change in a scripted workspace. Review the evidence yourself; these exercises do not launch vendor software or call live models.
Write Python around a scripted model client: validate requests and structured outputs, correlate tool results and restrict execution. The notebook runs local fixtures, not a live provider or deployment.
Each project starts from a fictional company's brief and moves between tools the way the job does: what you build in one stage is where the next one starts. Every brief is free to read.
Velomint
GtGitK8Docker and KubernetesAWAWS
Project
Recover from a failed release
Release 2.14.0 broke one bike unlock in three until on-call rolled it back. Revert the bad commit in Git, ship 2.14.1 on Kubernetes behind a readiness probe, add the CloudWatch alarm that would have caught it, and write the incident report.
A patient changed a number in an invoice link and saw a stranger's bill. Read the portal's access log in the shell to find what one session took, fix the endpoint so it checks who is asking, and write the incident note that decides who must be told.
When you finish the units, test yourself across the tools of the job at once: 10 practical tasks in 65 minutes, each in the tool it is about and checked automatically. The clock keeps running if you close the page.
Tool
Tasks
Minutes
Python for software engineers
3
23
Coding interview patterns
1
23
Git
3
11
System Design
3
9
How it is scored
Each tool
Ready when you get two thirds of its tasks right or more, Close at one third, otherwise Not yet. A coding task earns the share of its tests that pass.
Ready for interviews
75% or more overall, and every tool at Ready.
Nearly ready
50% or more overall, and no tool at Not yet.
Keep practicing
Anything else. The report lists every task you missed with the lesson that teaches it.
The assessment is free. The result is practice feedback in the Interview Hub's report format, not a hiring decision. Retakes draw different tasks where the question pools allow.
The interview
The Interview Hub runs the software engineer interview round by round, at a fictional employer: you answer out loud or in writing, the interviewer follows up, and the report quotes your answers.
The first unit of every course, with no account: 5 of the 36 units on this pathway. Every project brief is free to read, and the job-readiness assessment is free.
One Course
One course in full. On this pathway that is one of these: Python (6 units), Git (6 units), Coding interview patterns (8 units), Python for software engineers (4 units), PostgreSQL for engineers (3 units), System Design (5 units), Coding with AI agents (1 units), Building with language models (3 units).
All Access
The whole pathway: every unit, every project stage and spoken interview practice.
A software engineer builds and runs the code behind a product: an API that takes bookings, a job that sends reminders, the page a customer pays on. The day is part writing code and part reading other people’s code, reviewing changes, fixing what broke and deciding how the next feature should fit. Teams care that your code works, that others can change it, and that you can explain your choices.
From one language to system design
You start with one language because every later step needs it. Git comes early because you will use it every day from your first week. The coding pattern units come before the deeper Python because that is what the first interview rounds test. System design builds on those skills. The next step adds coding-agent instructions and review. The final step builds request, output and tool boundaries around a model client.
Showing what you can do
The projects put you in the middle of real engineering work at a fictional company. At Velomint a release broke bike rental payments: you revert the bad commit in Git, ship the fix on Kubernetes behind a readiness probe and add the alarm that would have caught it. At Orchard Row you read an access log in the shell and fix an endpoint that showed one patient another patient’s invoice. Then the assessment tells you which area to practice before you apply.
Practice C# and .NET to compare typed collections, asynchronous services and resource ownership. The .NET interview guide explains CLR memory, LINQ, DI, EF Core and boundary tests before the independent C# rounds.
Common questions
Do I need a computer science degree?
Some employers ask for one and many don't; the job ads you read will tell you. The interviews test what you can do: write working code, explain it and design a system. This pathway practices exactly that, but it is practice, not a qualification.
The lessons on this pathway add up to about 50 hours. The coding pattern units take longer than their count suggests, because each problem runs against tests until it passes. We don't promise how long getting hired takes.
What does it cost?
The first unit of every course is free, with no account. One Course opens one course in full; All Access opens the whole pathway. The pricing page shows every plan and its price.