current
Amazon
Software Development Engineer II · Hyderabad
- Designing and launching a worldwide Bedrock + RAG content validation service — cut review time by 50%, iterations by 70%, with >90% accuracy against human review.
I'm Abhinav Kumar Singh, a software engineer at Amazon. Four years of shipping backend services, security dashboards, and AI products — now sharing the system design and interview craft behind it all.
fig. 01 — this site, as a high-level design. nodes are clickable.
I graduated from Thapar Institute of Engineering and Technology in 2022 with a degree in Computer Engineering, and got my first taste of production code as an intern at Samsung SDS — debugging a sprawling service library and learning that real software is mostly about reading code, not writing it.
At F5 Networks I lived in the world of application security: leading the UI for a global efficacy platform, building dashboards that made WAF policies and L7 DoS attacks legible, and shipping features to a BIG-IP platform serving 30 million users.
At ServiceNow I went deep on platform engineering and AI — helping build a service designed to replace Ansible entirely, and designing a RAG-based chatbot that turned hours of SRE back-and-forth into seconds of self-service answers.
Today I'm an SDE II at Amazon, building an AI-powered content validation service used worldwide. Along the way I've solved 500+ problems across coding platforms and sat on both sides of many interview tables — which is exactly what the rest of this site is about.
Software Development Engineer II · Hyderabad
Software Engineer II · Hyderabad
Software Engineer II · Hyderabad
Software Engineer I · Hyderabad
Software Engineer Intern · Gurgaon
A structured course on high-level and low-level design — taught the way I'd answer in a real interview. Every problem walks the same six-step framework, so the approach becomes muscle memory before the whiteboard does.
What the system must do. Scope ruthlessly — interviews are won here.
Scale, latency, availability, consistency. The numbers that shape every choice after.
The nouns of the system — the data model everything else hangs off.
Contracts between components. Endpoints, payloads, and idempotency.
The boxes-and-arrows diagram — services, queues, caches, and data flow.
Bottlenecks, failure modes, and trade-offs. Where senior signals live.
first problem in progress: design a distributed logging system — kafka, elasticsearch, s3, redis.
A real coding round: one to two problems, follow-ups on complexity and edge cases, and honest feedback on how you communicate while you code.
60 min · scheduling opens soonA full system design interview using the six-step framework — requirements to deep dives — with a written scorecard on structure, trade-offs, and depth.
60 min · scheduling opens soonObject modeling and design patterns under interview pressure — class diagrams, interfaces, and extensibility questions, the way top companies ask them.
45 min · scheduling opens soonWrite-up 01
What a backend loop at a big tech company actually tests — round by round.
Write-up 02
The DSA patterns that kept showing up across my interviews — and how I prepared.
Write-up 03
System design rounds: what separates a pass from a strong hire.