AI Should Make You Smarter, Not Lazier
Why I built ReadingBuddy. Thoughts on privacy in the age of AI, using AI as a thinking partner instead of a crutch, and why your learning process is too personal to hand over to corporations.
Hey, I'mAman Khalid
I build systems that scale.
Software Engineer at Expedia Group specializing in distributed systems and large-scale migrations. Previously researched internet measurement and network infrastructure at Northwestern University.
01. About Me
I'm a software engineer who likes building things that work at scale. My background is a bit unconventional - I started in academia researching how the internet actually works, then moved to industry to build systems that millions of people use every day.
At Northwestern, I ran measurement campaigns across 33 cloud instances in 18 countries to understand how major cloud providers peer with each other and how submarine cables affect global connectivity. It sounds nerdy because it is - but it gave me a deep understanding of distributed systems that most engineers never get to see.
Now at Expedia, I work on large-scale service migrations and infrastructure. I've led projects that saved $114k annually in infrastructure costs and improved test coverage from 52% to 85%. I also got picked to lead our team's adoption of AI coding tools, which has been pretty fun.
Here are some technologies I work with regularly:
- Python
- Java / Kotlin
- Kubernetes
- gRPC
- AWS / Azure
- Distributed Systems
02. Where I've Worked
Software Development Engineer II @ Expedia Group
Feb 2023 - Present
- Led a large-scale migration from legacy monolithic services to modern distributed APIs, affecting millions of daily requests across Vrbo, Hotels.com, and Expedia.com
- Designed and implemented gRPC service integrations to unblock critical migration dependencies
- Reduced annual infrastructure costs by 63% ($114k savings) through strategic logging optimization
- Drove test coverage from 52% to 85% by implementing Jacoco verification thresholds
- Selected as early adopter and technical lead for AI coding tools initiative - created deployment guides and delivered presentations across the organization
Graduate Researcher @ Northwestern University
Sep 2021 - Dec 2022
- Designed and executed internet measurement campaigns using 33 cloud instances across 18 geographic locations
- Built a Python data pipeline processing 179 daily measurements from 358 instances worldwide
- Operated a global submarine cable network outage monitoring platform for real-time anomaly detection
- Analyzed BGP routing patterns and ISP peering relationships to understand global internet topology
Research Assistant @ Saint Louis University
Oct 2019 - May 2021
- Built a real-time ultrasound image processing pipeline supporting 100 concurrent streaming sessions
- Implemented using Kubernetes orchestration with sub-100ms latency requirements
- Published research at IEEE NFV-SDN 2020 on optimization techniques for distributed systems
Backend Software Engineer @ Marsplay
May 2017 - Jul 2019
- Architected microservices-based backend infrastructure handling thousands of daily transactions
- Built internal CRM tools that improved user onboarding efficiency by 45%
- Designed REST APIs facilitating communication between internal services and external integrations
03. Things I've Built
Featured Project
Global Internet Measurement Platform
A distributed measurement infrastructure spanning 18 countries that analyzed BGP routing patterns, ISP peering relationships, and submarine cable connectivity. Processed millions of traceroutes to understand how the internet actually works at a global scale.
- Python
- AWS
- Azure
- BGP
- Data Pipelines
Featured Project
ShedboxAI Agent
An open-source conversational AI agent that turns natural language into data workflows. Features a multi-tier memory system, specialist agents for complex tasks, and self-healing YAML generation. Built with a code-execution-first approach instead of traditional tool calling.
- Python
- LiteLLM
- ChromaDB
- Pydantic
- asyncio
04. Writing
Microservices Are Making Engineers Dumber
A pragmatic look at why the industry's favorite architecture pattern might be solving problems you don't have, while creating ones you definitely didn't need.
What I Learned Building an AI Agent From Scratch
Lessons from building ShedboxAI Agent. Why code execution beats tool calling, why small specialists beat one big agent, and why context management is the whole game.
What I Learned Leading Our Team's AI Tooling Adoption
Getting picked to lead our team's Claude Code adoption was equal parts exciting and terrifying. Here's what actually worked, what didn't, and why I think AI pair programming is here to stay.
Migrating Millions of Requests Without Anyone Noticing
Large-scale migrations are terrifying. Here's the playbook we used at Expedia to move millions of daily requests from a legacy monolith to a modern distributed architecture without breaking anything.
What Running Internet Measurements Taught Me About Distributed Systems
I spent a year at Northwestern sending packets across the globe and watching what happened. Turns out the internet is way more fragile and fascinating than most engineers realize.
How I Made a Service Like S3 Over the Weekend
Sometimes the best way to understand something is to build it yourself. Here's my weekend adventure in distributed object storage.
Best Practices for Building a Large Scale React Application
Lessons learned from building and maintaining React apps that actually scale. Spoiler: most "best practices" articles miss the hard parts.
Creating Awesome SPAs with React
A practical guide to building single-page applications that don't suck. Routing, state management, and all the gotchas nobody tells you about.
05. Get In Touch
I'm always open to chatting about interesting problems, new opportunities, or just nerding out about distributed systems and internet infrastructure. My inbox is always open.
Say Hello