Artificial Intelligence | Machine Learning | Software Engineering

Projects

Agentic AI for Observability & Control

Built end-to-end observability platform for BMC using agentic LLM. Integrated FastAPI, MongoDB, Prometheus, Grafana, LangChain, and Gemini 1.5 to monitor server telemetry and trigger self-healing actions via natural language.

TechStack: LangChain FastAPI MongoDB Prometheus Grafana React

Human Activity Recognition (HAR) on MMASH

Trained machine learning and deep learning models for human activity recognition using MultiModal Activities for Stress and Health dataset. Classified user activities based on sensor data and evaluated generalization of models using cross-validation.

TechStack: Python TensorFlow Scikit-Learn Keras

DDoS Detection using Reinforcement Learning

Built an intrusion detection system using Deep Q-Networks in a custom OpenAI Gym environment on the NSL-KDD dataset specialized to detect DDoS attacks. Designed and implemented a risk-weighted reward function to minimize false positives.

TechStack: Python PyTorch Jupyter OpenAI Gym

Bob The Agent

Multi-Agentic construction management network using Crew AI agents and Masumi payment integration. Making the process of bidding, scheduling, payments, and progress tracking more efficient, transparent, and secure.

TechStack: Python CrewAI Cardano

Personality test using NLP

Leverages Natural Language Processing and Machine Learning to analyze user inputs and predict user personalities with characters from "Game of Thrones" TV series integrating zero-shot learning for high-accuracy personality predictions.

TechStack: Python TensorFlow HuggingFace Scikit-Learn

Real Time Weather API

Web API that provides real-time weather information based on user input or IP geolocation. It includes secure authentication using JWT and integrates with MongoDB for persistent user data management.

TechStack: .NET Core MongoDB REST API JWT

About Me

Udayan Atreya

I am a software engineer with over three years of professional experience at Accenture, specializing in building scalable systems using AWS, CI/CD pipelines, and REST APIs.
Currently, I am pursuing a Master’s degree in Computer Science at San Jose State University, focusing on generative AI, prompt engineering, natural language processing, and cloud technologies.

Professional Timeline

May 2025 – Aug 2025

ML Teaching Assistant

Cornell University

  • Mentored 60+ students in ML, deep learning, and real-world coding best practices.
  • Guided teams in hyperparameter tuning and feature engineering, helping 80% secure internships or jobs.
Cornell
SJSU

Aug 2024 – May 2026

Graduate Student

San Jose State University

  • Pursuing MS in Computer Science with a focus on AI and cloud technologies.
  • Engaged in research and coursework in generative AI, NLP, and distributed systems.
  • Coursework: Retrieval Augmented Generation (RAG), Applied Deep Learning, Biometric Security with AI.

Dec 2021 – Jul 2024

Software Engineer

Accenture

  • Led development of RESTful microservice harnessing .NET, AWS Lambda and API Gateway enabling real-time user notifications during remediation with database persistence; scaling up by 20% user base.
  • Engineered custom CI/CD task orchestrating 5+ microservices for remediating code vulnerabilities; harnessing AWS serverless architecture and .NET; safeguarding 400+ applications over 3+ orgs.
  • Decreased manual ops labour up to 60% by enabling chatbot integrations for 100+ services; developed modules, orchestration flows and led cross-functional collaboration to onboard 20+ teams seamlessly.
Accenture
Accenture

Nov 2020 – Nov 2021

Associate Software Engineer

Accenture

  • Optimized legacy PowerShell scripts to start 20+ services concurrently by batching per server; reducing latency from 5 min to 1 min per operation, achieving 70%-time savings for stakeholders.
  • Reduced remediation latency by 37% through DynamoDB schema and CRUD service optimizations, eliminating data redundancy and streamlined look ups during root cause analysis.

Dec 2019 – Apr 2020

Software Intern

Motherson Tech

  • Constructed web application dashboard on Java Spring-MVC framework to visualize 50k+ tax receipts stored in PostgreSQL database.
  • Integrated data visualization libraries amplifying ticket resolution efficiency by 14% for stakeholders.
Motherson Tech
UCWL

May 2019 – Jul 2019

Network Software Intern

UCWL

  • Developed a custom .NET solution automated to run on scheduler to email report file containing health & running status of 300+ IP devices present in multiple factories.
  • Authored 3+ network topologies for factories allowing quick resolution reducing downtime by 20%.

Aug 2016 – May 2020

Undergrad Student

Manipal University Jaipur

  • Bachelor of Technology in Computer Science & Engineering
  • Coursework: Data Structure and Algorithms, Relational Database Management Systems, OOPS
MUJ

Contact Me

Social