Practice & Challenges

A typing speed test on a Turkish word list, saving results so you can track progress. Unlike English-centric tests, it measures with Turkish characters and word frequencies.

One puzzle a day each December, with the archive open year round. Unlike other puzzle sites everyone solves in sync, so the discussion and solution comparisons that week are unusually rich.

Java implementations of algorithms and data structures. Unlike textbook pseudocode, each is real, readable, working code with its tests.

A curated problem set with a video walkthrough for each. What the price buys is finishing a bounded path instead of getting lost among LeetCode’s thousands.

A stepwise framework for approaching an interview algorithm problem — constraints, examples, brute force, optimise, test. Method instead of panic.

A Japan-based contest platform whose beginner rounds (ABC) are more approachable for newcomers than Codeforces.

A resource list for the practice of code review. Unlike tool lists, it collects not only review tools but also articles and guidelines on review culture.

An online judge grading programming, mathematics and SQL problems automatically, formerly URI Online Judge. Unlike LeetCode it is learning- rather than interview-oriented, with plenty of beginner problems.

A single colour-coded table of time and space complexity for data structures and sorting algorithms. Unlike a textbook it shows everything at a glance — the last look before an interview.

A piece correcting common misunderstandings of big-O notation. Unlike standard explanations, it targets subtle errors directly, such as conflating worst case with average case.

A collection of common bit-level tricks — the patterns that turn up in low-level optimisation and interview questions.

In-browser lessons across 15+ languages. Because everything runs in the browser instead of a local setup, it removes the barrier that stops beginners at installation.

A community contest to solve a problem in the shortest possible code. Unlike algorithm sites the measure is code length rather than speed or correctness — it shows how far languages can be bent.

A problem sequence starting at the very easiest and climbing gradually — a far gentler curve than the contest platforms for absolute beginners.

A competitive algorithm platform with regular contests, difficulty-tagged problems and published editorial solutions. Unlike Codeforces, it offers learning paths and practice sections aimed at beginners.

The busiest competitive programming platform, with regular rounds and a rating system. Problems run harder than LeetCode and lean heavily mathematical.

A practice-heavy course teaching Java through 1,200 hands-on tasks, with every solution checked automatically. Unlike video courses, it is tasks rather than theory: you advance by writing code.

Front-end (HTML/CSS/JS) challenges with an embedded editor, each round on a new theme. Unlike algorithm sites the result is visual: there is no single right answer, just creative solutions.

An interview problem bank and assessment tool. Unlike LeetCode it offers video explanations of solutions and also hosts the assessment tool companies use to screen candidates.

A skills assessment and interview platform that produces a standardised score you can send to several companies. The single assessment is the difference from sitting a separate test for every employer.

Practice gamified with martial-arts ranks. Unlike LeetCode the learning happens after you solve — you read and vote on what everyone else wrote for the same problem.

The assessment platform companies use for technical screening, with a practice section open to candidates. Its difference from other problem sites is that it mirrors the format of real screening tests — you see the exam environment before the exam.

A platform that has you solve programming puzzles with a game theme, showing your code’s result in a game scene. Unlike dry problem sets it animates your solution visually.

RPG-themed learning that sustains motivation with XP and map progress. Unlike serious-toned courses it leans on the game theme; built for complete beginners, it does not carry you to advanced.

A free handbook for competitive programming, from basic algorithms to advanced techniques. Against academic algorithm books it is contest-oriented: it hands you an implementable template rather than a proof.

An English translation and extension of the e-maxx.ru write-ups on competitive-programming algorithms, each topic with code and practice problems. Unlike problem-solving platforms it explains the algorithm itself in depth rather than serving problems.

Teaches cryptography by breaking it, with each level exploiting a real weakness. Unlike a theory course it teaches from the attacker’s side — the applied counterpart.

A visualisation tool built for Georgia Tech’s data structures course — narrower than VisuAlgo but aligned exactly to a syllabus.

A service that emails one coding-interview problem every day. Unlike problem platforms it lands in your inbox — the point is building a habit; detailed solutions are paid.

SQL and data science interview questions with real queries run and checked in the browser. Unlike general algorithm platforms, it focuses on the data interviews they leave uncovered.

A challenge site that has you build web projects with realistic designs and requirements. Unlike random projects the community reviews the solutions.

Small, fast puzzles targeting language fluency rather than algorithm theory — better than LeetCode for settling syntax when you move to a new language.

Exercise-based practice in 70+ languages, with volunteer mentors commenting on your solution. Unlike LeetCode-style sites, the goal is idiomatic use of the language, not speed or interviews.

Language-independent definitions and test data for Exercism’s exercises. Unlike the Exercism site it is raw material rather than a practice environment — for building a practice set in your own language.

A complete free curriculum with in-browser exercises and free certificates. Its capstone projects make you build real applications rather than stopping at drills.

A very broad article archive of algorithms, data structures and interview questions. Unlike single-topic sites it has a worked example for almost every CS topic.

Run Git commands in a safe environment with immediate feedback. Rather than experimenting on a real repository, you practise here without the fear of breaking anything.

A practice tool where you answer interview questions aloud and get AI feedback. Unlike coding-interview sites, it focuses on behavioural and role questions, highlighting the key terms in your answer.

Combines contests with hiring assessments — companies run technical screening here. Unlike HackerRank, it is strong on hosting hackathons.

Practice and hiring assessments on one platform. Unlike LeetCode, companies genuinely run their technical tests here, so you get to know the exam environment in advance.

Structured courses for system-design and algorithm interviews. Unlike problem banks it teaches method rather than questions — a framework for working a problem step by step; the content is old but the framework holds.

Opens up the reasoning steps rather than handing over the answer, showing how you move from brute force to optimal — which is exactly the reasoning an interview looks for.

Gamified interview preparation. Unlike LeetCode’s open problem pool, topics are ordered by dependency, each building on the last.

A list of questions to ask the company in an interview. Unlike candidate-prep lists it flips the direction, so you assess the workplace.

Touch-typing practice that measures which letters you struggle with and generates text accordingly. Against fixed-text typing tests that adaptation is the point — your weak letters keep coming back.

Spins up a real Linux or Kubernetes machine in the browser — not a simulation. Its CKA and CKS scenarios sit very close to the actual exam environment.

Draws the commit graph live to show what rebase, cherry-pick and merge actually do — the shortest path to learning Git by understanding the graph rather than memorising commands.

A practice archive focused on JavaScript interview questions: writing polyfills, the event loop, closures. Unlike general algorithm sites it is language-specific — "implement bind yourself" is not a question you find elsewhere.

The most widely used bank of interview questions, filterable by company tag. The discussion section is usually more instructive than the official solution.

Tushar Roy’s large collection of coding-interview problems and solutions. Unlike LeetCode solutions, most problems pair with a detailed video explanation on YouTube.

A roadmap that groups LeetCode problems by topic and pairs each with a solution video. Against a raw problem list the difference is ordering: which 150 problems, in which sequence, teach the patterns — the thing random grinding never gives you.

Turkish touch-typing training supporting F and Q layouts separately. Unlike Q-centric sites it supports the F layout too — free F-layout material is rare locally.

Free, peer-to-peer mock technical-interview practice, now part of Exponent. Unlike solving problems alone there is a real person opposite you, and you take turns as interviewer and candidate.

Teaches programming through many small exercises, designed for a high-school-level start. Unlike project-based courses, each step adds only a little to the last.

Mathematical problems where brute force does not finish; solutions need number theory or combinatorial insight. Not to be confused with interview questions — it trains a different muscle. Its archives page lists every problem with the number of people who solved it.

Well-explained small programming projects, heavy on maths and algorithms. Unlike problem sites, each article comes with a clean implementation, often in several languages, alongside the explanation.

A riddle game where each level is solved with a bit of Python. Unlike exercise sites every puzzle hides the next level’s address — you cannot move on without solving it.

A repository that teaches Python by making failing tests (“koans”) pass one at a time. Unlike reading a tutorial you fix real tests, so feedback is immediate.

Steps through code drawing stack frames and heap objects. Unlike a classic debugger it visualises memory — nothing explains reference semantics, scope and mutable default arguments as clearly.

Peter Norvig’s short Python programs and notebooks on puzzles, algorithms and language. Unlike exercise sites each is an expert’s worked solution with the reasoning explained.

A spec whose same realistic app (Conduit) is built across different front-end and back-end stacks.

A wiki showing the same small problem solved in hundreds of languages. Unlike single-language examples it shows side by side how a task differs across languages.

An organised, interactive space of topic tutorials and practice for aspiring programmers. Unlike Scaler’s paid academy programmes, these are free, standalone topic pages.

A scenario-based game for learning SELinux that you run on your own machine. It turns the usually dry topic of policy writing into practical debugging tasks. Unlike plain documentation, it teaches policy through debugging tasks.

A Python 3 course with 100+ exercises, coded in the browser. Unlike video courses a checker verifies your solution at once.

A long-running online judge with a large archive of classic algorithm problems. Unlike Codeforces it is more archive than contest and accepts 40+ languages; the interface is dated, but the problem set is still a reference.

Stanford’s problem-solving course for technical interviews with open materials — it teaches a way of thinking rather than a question bank.

A channel presenting data structures and algorithms as an ordered curriculum. Unlike scattered solution videos it is built to be watched in sequence.

An ordered, comprehensive, free guide to cracking the tech interview, by the author of the Blind 75 list. Unlike problem banks it covers the whole process — résumé, behavioural questions, salary negotiation.

An archive of data structure and algorithm problems grouped by topic, with C++, Java and Python solutions. Against LeetCode the solution arrives with its explanation, which makes it readable before you attempt the problem.

A platform teaching web development through projects, with a full open curriculum. Unlike video courses it has you set up your own environment and build real projects.

Shows the same to-do app built in dozens of JavaScript frameworks. Unlike the frameworks’ own demos the functionality is held constant, so only the approach changes and you can compare them side by side.

One of the oldest competitive programming platforms; its problem archive and editorials form a slice of algorithm history. Unlike Codeforces, its weight has since shifted to crowdsourced freelance work.

The classic arena for timed algorithm competitions (SRM). Unlike Codeforces it uses the SRM format: after the coding phase comes a challenge phase where you score by feeding failing inputs to other people’s code.

The classic competitive programming tutorials, many from the early 2000s, on subjects that do not age. Against newer material the difference is depth: long pieces that work a single technique from start to finish.

A channel working algorithm problems step by step, best known for its dynamic programming series. Against reading a solution it shows the thinking aloud, which is what an interview asks for.

An online game to increase your typing speed by racing against others. Unlike solo typing tests it is a live race, with texts drawn from book and film quotes.

A classic online judge auto-grading hundreds of contest problems. Unlike newer platforms, it is the archive the problem lists of the Competitive Programming (Halim) book are built on.

Teaches Vim shortcuts through a Zelda-like game — you earn hjkl and motion commands by playing rather than memorising them.

Animates data structures and algorithms step by step. Unlike textbook diagrams you supply your own input and watch — something like an AVL rotation is far easier to grasp than by reading code.