find the gap
A 10-minute skill check maps hash maps, two pointers, sliding window and stacks, and names the skill you're actually missing.
// data structures & algorithms, one skill at a time
Most DSA prep is grinding problems and hoping the pattern sticks. Awelgorithms breaks DSA into 600+ concrete skills, finds the ones you're missing, and checks your code until each one is yours.
free · no sign-up for the check · about 10 minutes
next skill fixed-size windows
unlocks Maximum Average Subarray I on LeetCode
// try one now
This is what every question in the skill check is like: one decision, one concrete skill behind it. Answer it and you'll see the skill it tests and the LeetCode problems it unlocks.
No sign-up. Nothing is saved.
Sorted array, one pointer at each end, looking for a pair that adds up to the target. The pair sum is too small. Which pointer moves?
12 · sum 1 + 9 = 10 too smallPick one. You'll see the pointers move.
// why grinding stalls
Solving a problem proves you solved that problem. Interviews ask the next one. What transfers is the move underneath: which pointer to advance, when a window has to shrink, what the map should remember.
Awelgorithms teaches those moves one at a time, checks that you can use them, and only then sends you to the problems that need them.
| what you need | grinding LeetCode | curated lists | awelgorithms |
|---|---|---|---|
| tells you what you're missing | no: you find out mid-interview | by topic, not by skill | skill by skill, from a 10-min check |
| teaches the move before the problem | editorial after you're stuck | a walkthrough per problem | a worked trace per skill |
| checks your reasoning, not just output | tests pass or fail | no | graded on core and edge cases |
| knows when you're ready for a problem | no | a fixed order for everyone | unlocks problems as skills land |
| size of the problem bank | thousands | 75 to 150 picks | sends you to LeetCode for the reps |
Curated lists means Blind 75, NeetCode 150 and the like. Use them for coverage; use Awelgorithms for the skills they assume.
A 10-minute skill check maps hash maps, two pointers, sliding window and stacks, and names the skill you're actually missing.
Each skill is one concrete move, taught with worked traces you can follow line by line.
Write the code yourself. Awel checks it against core and edge cases, then points you at the LeetCode problems it unlocks.
// inside a lesson
This is the real lesson behind “fixed-size windows”: the core idea, a worked trace, and the drill you write next.
> open this lessonk is given. Every window has length k. For n cells there are n - k + 1 windows.
a = [1, 2, 3, 4], k = 2[0,1] [1,2] [2,3]4 - 2 + 1 = 3 windowsyour turn write windowCount(n, k)
16 questions, no sign-up. A shareable skill map and your next skill.
Not "arrays", but "two pointers from both ends". 36 tracks, from arrays to shortest paths.
Your code, in any of 18 languages, checked by Awel against the skill's core and edge criteria.
Ask Awel why it works or where your reasoning broke, right next to the drill.
// patterns
How to spot each one, the shape of the code, the mistakes that cost interviews, and the problems that use it.
A hash map trades memory for time: remember what you have seen, keyed by the thing you will look up later, and a nested loop becomes one pass.
3 problemsA hash set answers one question fast: have I seen this before? Use it when you only care whether a value exists, not how often or where.
3 problemsMany array problems are solved in one pass by carrying a little state forward instead of looking back: the best value so far, a running minimum, a write index, or prefix and suffix products.
5 problemsTwo pointers walk the input from two places at once, usually the two ends of a sorted array or string, and each step rules out a whole set of answers.
4 problemsA sliding window keeps a contiguous range and its state (a sum, counts, distinct values) and updates them as the range moves, instead of recomputing every subarray.
4 problems90 lessons across hash maps, two pointers, sliding window and stacks.
browse// founding cohort
Prep with 14 other people. 4 weeks, free, starting late October or early November.
Most DSA practice measures whether you solved a problem, not whether you could solve the next one. I built Awelgorithms to fix that: name the move, practise it until it's yours, then go solve problems.
The founding cohort is how I find out whether it works. I read every application and run the weekly check-ins myself.
// faq
Yes. The skill check, the roadmap, the lessons, the graded drills and tutoring with Awel are free. Paid plans with more time with Awel will come later; the free plan stays the main product.
No. It's 16 questions and takes about 10 minutes. You get a skill map you can share and the next skill to learn. Create an account only if you want to save progress.
hash maps, two pointers, sliding window and stacks: the patterns behind most interview questions. Each answer maps to one concrete skill, so the result names what to practise, not just a score.
LeetCode is a great problem bank, and we send you there. What it doesn't do is tell you which skill you're missing or teach it. Awelgorithms breaks each pattern into small skills, teaches each one with a worked trace, checks your code, and unlocks the LeetCode problems you're ready for.
18 languages, including Python, Java, C++, JavaScript, TypeScript, Go and Rust.
Awel is the AI tutor inside every lesson. It knows which skill you're on, checks your drill against that skill's core and edge cases, and answers questions like why a step works or where your reasoning broke.
A free 4-week cohort for 15 people with a coding interview coming up. You start with the skill check, practise one skill a day, and have a weekly check-in with the founder.
Ten minutes, sixteen questions, one honest skill map.