Sorted Arrays & Palindromes
Used for sorted pairs (Target Sum), container with most water, Dutch National Flag, and palindrome verification in $O(N)$ time with $O(1)$ extra space.
Engineering Reference
Software Engineering Interview Cheat Sheet & Frameworks
A reference guide for L4, L5, and L6 engineering loops: Big-O time and space bounds, essential algorithmic patterns, 15-stage system design blueprints, and capacity math formulas.
Part 01 · Algorithms
Top 8 Algorithmic Patterns for Technical Loops
Master these core templates to solve 90%+ of LeetCode Medium and Hard problems.
Substrings & Contiguous Subarrays
Dynamic and fixed window sizes for longest substring without repeating characters, minimum window substring, and max consecutive ones in $O(N)$ time.
Linked List Cycles
Floyd's Cycle Detection algorithm for detecting loops in linked lists, finding middle nodes, and identifying the start of a cycle in $O(1)$ space.
Next Greater Element
Used for daily temperatures, largest rectangle in histogram, and sliding window maximum in linear $O(N)$ time.
Part 02 · Numbers
Back-of-the-Envelope Capacity Numbers Every SWE Must Know
Standard engineering constants for system design estimation rounds.
| Metric / Operation | Scale / Latency | Engineering Context |
|---|---|---|
| 1 Day in Seconds | 86,400s (~100k) | 100M daily requests = ~1,200 avg QPS (~2,400 peak) |
| RAM Access Latency | ~100 ns | Redis / Memcached cache lookups |
| NVMe SSD Random Read | ~100 µs | Indexed relational database query lookup |
| Cross-Datacenter Network | ~50 – 150 ms | Multi-region asynchronous replication budget |
SWE FAQ
Frequently Asked Questions on SWE Interview Prep
Tactical advice on pacing, frameworks, and mock interview practice.
What are the core technical patterns tested in software engineering interviews?
Software engineering technical loops evaluate 4 major patterns: Data Structures & Algorithms (Two Pointers, Sliding Window, Monotonic Stack, Dynamic Programming), System Design (15-stage multi-tier architecture), Low-Level Design / OOP (Design Patterns, Concurrency), and STAR Behavioral Fit.
How does ClawPad help during software engineering mock practice?
ClawPad streams structured algorithmic talk tracks, 15-stage distributed system design blueprints with real-time SVG diagrams, and a capacity model whose arithmetic the desktop re-checks for consistency before rendering.
What are the key capacity planning numbers software engineers must know?
1 day = 86,400 seconds (~100k for estimation). 100 QPS = ~8.6M requests/day. 1 Million requests of 1 KB = 1 GB. L1 Cache reference ~0.5ns, RAM ~100ns, SSD ~100µs, Datacenter roundtrip ~500µs, Cross-country roundtrip ~150ms.
Practice with live streaming guidance
Download ClawPad