Software Engineer based in Seattle, Washington.
Rishab Katta
Software Engineer based in Seattle, Washington.
About Me
Hey, I’m Rishab.
I’m a Software Engineer with a Master’s in Computer Science from Rochester Institute of Technology, and I’m drawn to building systems that sit quietly in the background, doing complex things reliably, and making large-scale human workflows simpler and fairer.
I spent the last few years at Amazon, working in the Last Mile org on Amazon Flex — a large-scale logistics platform that powers on-demand deliveries through independent delivery partners across the world. Flex operates at massive scale, and much of the work lives deep in backend and distributed systems, where small architectural decisions can have outsized real-world impact.
On the Flex team, I worked primarily on core platform systems that determine how delivery offers are generated, validated, and served to drivers in real time. My work focused on problems like:
- Redesigning and migrating critical DynamoDB data models that sat on the hot path of the system.
- Building distributed services that precompute and cache complex business logic to reduce latency and improve reliability.
- Designing systems to handle fairness, compliance, and safety — including mechanisms to prevent overwork and enforce rest periods.
- Launching new offer experiences and platform capabilities across multiple countries, working closely with product, operations, and compliance teams.
A lot of the work I enjoyed most was invisible to end users: fixing race conditions, untangling legacy architectures, improving system reliability, and reducing latency by orders of magnitude. These weren’t flashy features, but they directly shaped how millions of people interacted with the platform every day.
What I liked about Flex was that it sat at an interesting intersection of infrastructure and human behavior — algorithms deciding who gets which work, systems enforcing fairness, and real people on the other side whose time, income, and experience were affected by these decisions.
Before Amazon, I worked as a Software Engineer at Infosys, building backend services for financial systems, focusing on server-side business logic, multi-threaded applications, and large enterprise platforms.
During my Master’s, I also interned at Nokia and Aeris Communications, where I worked on cloud infrastructure, secret management systems, and an NLP-based internal support platform. At Aeris, I also contributed to research on firmware-over-the-air (FOTA) systems for IoT devices, studying how network constraints affect large-scale deployments.
My technical interests lie at the intersection of distributed systems, data platforms, infrastructure, and product engineering — especially systems that operate at scale, have real-world consequences, and require careful thinking about reliability, fairness, and long-term design.
Outside of engineering, I’m interested in street photography and visual storytelling, and I occasionally share photos on Instagram.
You can find my projects and experiments on GitHub:
github.com/rishabkatta
And some of my photography here:
instagram.com/notrishabmaybe
Work Experience
Software Engineer
Software Engineering Intern
Software Engineering Intern
Software Engineer for Amazon Flex - working on backend and distributed systems for a large-scale, real-time logistics platform that uses gig labor to deliver Amazon packages to customers fast and reliably.
Software Engineering Intern
Software Engineering Intern
Software Engineering Intern
Worked remotely for Nokia Software from Sept '20 - Dec '20. Built a Distributed Secrets Management Platform using GoLang and integrated it into Enterprise Nokia Software Product Portfolio.
Software Engineering Intern
Software Engineering Intern
Software Engineering Intern
Interned at Aeris Communications' San Jose Office from Aug'19-Jan'20. Built a chatbot from ground-up and worked on a research project & published a paper on FOTA transfer on LTE-M network.
Software Engineer
Software Engineering Intern
Software Engineering Intern
Worked as a full-time software engineer in Infosys's Bangalore Office from Jun'16 - Jun'18. Built a highly scalable check payments verification app using Java and Spring MVC Framework.
Academic Projects
NetCache - Fast In-Network Caching
NetCache - Fast In-Network Caching
NetCache - Fast In-Network Caching
Reproducibility Study on the paper published by Xinjin et al. Implemented the switch code in P4 language and developed a mininet for NetCache performance demonstration.
Kademlia Distributed Hash Table
NetCache - Fast In-Network Caching
NetCache - Fast In-Network Caching
State-of-the-art Distributed Hash Table implementation in Java. Exercise on P2P Systems.
Publisher/Subscriber Model
NetCache - Fast In-Network Caching
Diabetic Retinopathy Grading
Developed a Highly Scalable Publisher/ Subscriber Model with a beautiful Command Line Interface using Socket Programming, Remote Procedure Call (RPC) protocol in Java and using Docker Containers.
Diabetic Retinopathy Grading
Image Thresholding -Otsu's Method
Diabetic Retinopathy Grading
Developed a Robust Deep Learning model for Automated Diabetic Retinopathy Grading using Inception V3 Convolutional Neural Network, Stochastic Gradient Descent Optimization, Batch Normalization and Pre-initialized ImageNet weights using EyePACS dataset
Image Thresholding -Otsu's Method
Image Thresholding -Otsu's Method
Image Thresholding -Otsu's Method
Wrote an Unsupervised Learning Algorithm for clustering data into two clusters based on Otsu's method. Wrote a Mentor Program that creates a Trained Program that does classifications on the Validation dataset.
AI Agent for playing Berzerk
Image Thresholding -Otsu's Method
Image Thresholding -Otsu's Method
Programmed an AI Agent to achieve a high score on Atari Berzerk. The raw pixel data is analyzed and used to detect player movements, enemy firing and maze walls. Implemented A* Search Algorithm for finding the shortest path to the nearest exit.
Exploring Discourse in Reddit
Exploring Discourse in Reddit
Created an ML Agent that can predict which subreddit an unlabeled post comes from.SVM, RandomForest and LSTM models are created and their performances are compared using Precision Recall & ROC curves.
Stocks Performance Prediction
Exploring Discourse in Reddit
Applied Itemset Mining and Association Rule Mining algorithms on a large stocks dataset
to predict the percentage increase of stock prices and the companies whose stock prices increases together.
Donald Knuth Conjecture
Wrote Python Code for solving Donald Knuth Conjecture. Donald Knuth has conjectured that every positive integer can be obtained by beginning with a single 3 and applying some combination of the factorial, square root, and floor functions.
Decision Tree
Implemented a Decision Tree from scratch to predict if a person would wait for a table at a restaurant. This is a textbook problem from Artificial Intelligence: A Modern Approach, By Stuart Russell and Peter Norvig
Blood Supply-Demand Est.
Blood Supply-Demand Est.
Implemented Ford Fulkerson Algorithm on a multi-source, multi-sink directed graph to determine whether the current blood in reserve with the hospital is enough to meet the demand.
Personal Projects
Yelpcamp
Yelpcamp is a campgrounds sharing website where you can share and discover new campgrounds. You can Signup/login, Post/edit reviews, find location details. I built this app using Node.js, Express, MongoDB, Bootstrap 4 and RESTful Routing.
Salat
Salat is a minimalist b/w-photos-only photo sharing website. I love b/w photography and I built this app using HTML, CSS3, Bootstrap 4, Javascript and jQuery.
RCS
RCS - Rishab's Candy shop, is a huge, fictional, willy-wonka style chocolate shop for which you can buy tickets for a visit, participate in contests and win gift cards. I built this app using the MEAN stack.
Blogsite
Blogsite is a tech blogging website where users can login/signup and write/read/edit articles about the current trending technologies. Created using MEAN stack and RESTful Routing.
RK-Studio
RK-Studio is a controller emulator for music making. I built this app using paper.js for animations and howler.js for music samples.
To-Do List Web App
A simple to-do list web app created using HTML, CSS, Javascript, and jQuery. My Frontend Hello-World!