Pratikshit portrait

I build systems that don't flinch under scale—event-driven services, data pipelines, and GenAI apps that actually ship.

Currently focused on designing microservices that stay up, fast, and readable with production GenAI patterns (RAG, agents, evaluation). I've spent time at Intuit, Zeta Global, Illinois Center of Transportation, and JP Morgan Chase building systems where correctness, performance, and maintainability all matter.

Open to collaborating on: RAG/Agent systems solving real workflows, data platforms (PySpark/EMR, event-driven architectures), and backend systems where scale meets observability.

About Me

I am open to new opportunities for full-time positions as a Backend Engineer, Software Engineer, or AI Engineer. Feel free to reach out!

My core expertise lies in full-stack development, designing event-driven microservices, building scalable data pipelines, and shipping production GenAI systems. I'm proficient in Spring Boot, Python, Kafka, PostgreSQL, AWS, Docker, Kubernetes, LangChain, RAG patterns, PyTorch, React, Next.js, and TypeScript.

I'm currently exploring production GenAI patterns—evaluating RAG approaches, building reliable agents, and making LLM systems measurable and safe. I love problems where clean architecture, operational excellence, and Hexagonal design patterns intersect.

Open to Collaborate On:
• Retrieval-Augmented Generation (RAG) and Agent systems solving real workflows
• Data platforms (PySpark/EMR, streaming, event-driven architectures)
• Backend systems where performance, correctness, and maintainability all matter

When I'm not working, you can find me lifting or eating.

Education

university
University of Illinois Urbana-Champaign

Masters of Computer Science

university
Indian Institute of Technology Varanasi

B.Tech. in Electrical Engineering

My Experience

My Projects

Youtube Playlist Tracker

  • Kafka application for users to track their playlist added in the app
  • Uses Youtube API to fetch video statistics
  • Implements Kafka producer-consumer model to track video statistics
  • Sends notifications to users on Telegram
  • Kafka
  • Confluent Cloud
  • Heroku
  • Youtube API
  • Telegram API
  • Python
  • Flask
  • ksqlDB
View ProjectClick card to expand

Life Cycle/Cost Assessment Tool

  • Web app for computing cost and environmental impacts of pavement projects
  • Analysis period tracking and reporting for pavement projects
  • Django
  • React
  • Postgres
  • AWS Lightsail
View ProjectClick card to expand

Initial Pavement Rehabilitation

  • Web app for comparing cost of initial pavement rehabilitation using different treatments
  • Provides comprehensive analysis and recommendations for pavement projects
  • Django
  • React
  • Postgres
  • AWS Lightsail
View ProjectClick card to expand

Project Management Tool

  • Flask-based data analytics tool for ICT's project management
  • Users can upload project budget spreadsheets and generate spreadsheet reports
  • Saves numerous hours of manual work for the organization
  • Flask
  • HTML
  • CSS
  • XLSXWriter
  • Pandas
  • SQL
View ProjectClick card to expand

Fast Mutation Testing in Python

  • Fast Python mutation testing framework to validate test suite quality
  • Safely run mutation trials without source code modifications
  • Incremental code changes mutation generation for efficient testing
  • Mutation Testing
  • Python
  • AST
  • Mutatest
View ProjectClick card to expand

Attention-Enhanced Super-Resolution GAN

  • Implemented SRGAN with multi-scale attention U-net discriminator
  • Built residual-in-residual dense blocks generator
  • Achieved better NIQE score with sharp edge details and texture generation
  • Super Resolution
  • GAN
  • Attention blocks
  • RRDB
  • U-Net
  • CNN
View ProjectClick card to expand

Domain Adversarial Transfer Learning for Time Series Classification

  • Developed methods to generate superior pre-trained 1-D CNN transfer models
  • Optimized for time-series classification tasks
  • Domain adversarial networks
  • Error bounds
  • Timeseries
  • DTW
  • Divergence measures
  • CNN
View ProjectClick card to expand

My Skills

Backend & Microservices

  • Spring Boot
  • Django
  • Flask
  • Kafka
  • Redis
  • Nginx

Data & ETL

  • PySpark
  • EMR
  • Hadoop
  • Hive
  • SQL
  • Airflow

Cloud & DevOps

  • AWS
  • GCP
  • Docker
  • Kubernetes
  • Jenkins
  • Github Actions
  • ArgoCD
  • Terraform

GenAI & LLMs

  • LangChain
  • LangGraph
  • RAG
  • Vector DBs
  • Embeddings

ML/DL

  • PyTorch
  • TensorFlow
  • Keras
  • Scikit-Learn

Frontend

  • React
  • Redux
  • TypeScript
  • JavaScript
  • Tailwind
  • Next.js
  • HTML
  • CSS
  • Mantine

Databases

  • PostgreSQL
  • MongoDB
  • SQLite
  • GraphQL
  • Apollo

Languages

  • Python
  • Java
  • C++
  • C
  • Bash

Tools & Utilities

  • Git
  • Heroku
  • Vercel
  • Pandas
  • Numpy
  • Matplotlib

Contact Me

Please contact me directly at pratikshitsingh.jobs@gmail.com or through this form.