Built By Sid

Algo
Products

Grinding LeetCode never stuck for me. The patterns only clicked when I built something real around them. So that's what this is. 12 products, 12 patterns, one rule: do the thinking yourself. AI can help with everything else.

The Rule: the core algorithm gets built in a blank file. No AI.
12products to ship
01rule: no AI on the algorithm
languages — Go, Rust, Python
things to learn

The Why

Two birds,
one stone

I failed the last stage of a 6-stage interview because my algorithms weren't sharp enough. That stung. But more than the failure, it made me realise that grinding LeetCode problems in isolation wasn't going to fix it. The patterns weren't sticking.

There's also a bigger thing going on. AI is genuinely useful. But it's a force multiplier and it can only multiply something. If you stop doing the thinking yourself, there's nothing left to multiply. You need to keep that muscle working.

So this project solves both. Build something real around every pattern and the pattern sticks. Do the hard part yourself and the muscle stays sharp. Write about it and the thinking gets clearer. Every product is doing three things at once.

01

Build something real around every pattern

Abstract problems don't stick. An org chart explorer that runs BFS on The Office characters does.

02

One honest constraint

The core algorithm gets written in a blank file. No AI. Everything else is fair game. That one constraint is where the learning actually lives.

03

Write about it

Every product gets a dev log. Writing forces you to understand what you actually built. Shipping forces you to finish it.

04

Learning Go along the way

Go doesn't give you much for free. No built-in reverse, no built-in min. You write the primitives. That friction is part of the point.

The Catalogue

12 products,
ordered by interview relevance

01Go

Org Chart Explorer

BFS · DFS · Lowest Common Ancestor

02Go

Word Morph

BFS · Bidirectional BFS · A*

03Go

Kanban Snake

Stack · DAG · Topological Sort · Cycle Detection

04Rust

Link Vault

Trie · Union-Find · LRU Cache

05Go

Budget Pulse

Prefix Sums · Monotonic Stack · Binary Search

06Go

Type Racer Solo

Sliding Window · Two Pointers · Edit Distance DP

07Go

Focus Timer

Interval Merging · Greedy Scheduling · DP

08Go

Splitwise Lite

Greedy · Directed Graphs · Topological Sort

09Rust

Diff Tool

Edit Distance · LCS · Myers Diff

10Python

Sous Chef

Hash Maps · Sliding Window · BFS / DFS

11Rust

Pack It

0/1 Knapsack · Unbounded Knapsack · Multi-dim DP

12Go

DJ Mode

HashMap + Array Sync · Weighted Random · Stack

The Format

Every product,
same structure

The Technique

What the algorithm is and why it's hard. No hand-waving.

Algorithm Ladder

V1, V2, V3. Complexity grows with the product. Ship early, improve often.

The Constraint

The rule you honour to make the learning stick. No AI on the core algorithm.

The Self-Test

How you know your implementation is actually correct. Every product has one.

Follow Along

Dev logs on Substack

Every product gets a dev log. The thinking, the wrong turns, the bugs, the algorithm. Honest documentation of what it actually took to build it.

Read on Substack ↗