Hi, I'm Ashrith Vadde

AI/ML Engineer & Data Science Graduate Student

AI/ML Engineer with 2 years of experience developing and deploying LLM and RAG-based applications for enterprise and research use cases. Currently pursuing MS in Engineering Data Science and AI at University of Houston. Focused on translating complex AI concepts into scalable, production-ready solutions with real-world impact.

About Me

I'm a passionate AI/ML engineer and researcher currently pursuing my Master's in Engineering Data Science and AI at the University of Houston. My work focuses on developing intelligent systems powered by Large Language Models, Retrieval-Augmented Generation, and Deep Learning.

With professional experience at Accenture and research background at Samsung PRISM, I specialize in building production-ready AI solutions, explainable AI frameworks, and scalable ML pipelines. My research interests span Machine Learning, Generative AI, Explainable AI (XAI), Computer Vision, and Reinforcement Learning.

Skills & Technologies

AI & Machine Learning

Deep learning, LLMs, RAG systems, explainable AI (XAI), computer vision, and reinforcement learning

TensorFlow PyTorch OpenAI API LangChain SHAP OpenCV

Development & Frameworks

Full-stack development with focus on AI application deployment and REST APIs

Python Java FastAPI REST APIs SQL

Cloud & MLOps

Azure cloud services, ML pipeline automation, and model deployment expertise

Azure OpenAI Azure Databricks Git Postman

Featured Projects

GenLite - Code Conversion Platform

Enterprise LLM-based reverse engineering platform with function calling and advanced prompt engineering, reducing manual code migration effort by 70% for cross-language transformations.

LLMs Function Calling Prompt Engineering RAG
Proprietary project at Accenture

Corn Leaf Disease Classification

Developed a Deep ensemble learning model combining VGG16, MobileNet, InceptionV3, and Xception achieving 97.6% accuracy on corn leaf disease classification using data balancing and image augmentation.

TensorFlow Deep Learning CNNs Ensemble Methods

2048 Game - Deep Reinforcement Learning

Built a Deep Q-Learning agent from scratch in PyTorch to autonomously play the 2048 game, achieving highest tile of 1024 through reward shaping and iterative training.

PyTorch Reinforcement Learning Deep Q-Learning

Sentence Embedding Techniques for Summarization

Performed a comparative analysis of transformer-based and graph-based embeddings, evaluating semantic fidelity using ROUGE metrics to optimize text summarization performance.

Transformers Graph Embeddings ROUGE NLP

Publications

Playing Maze Using Voice Commands

ICCCNT 2023 • June 2023

Developed a deep-learning model that converts voice commands into spectrograms for real-time navigation. The system interprets spoken instructions to enable hands-free maze traversal through audio input processing and deep neural networks.

Deep Learning Audio Processing Spectrograms Real-time Systems

Real-Time Human-Computer Interaction Using Hand Gestures

ICTIS 2023 • April 2023

Created a computer-vision gesture interface enabling virtual mouse and keyboard control with real-time hand-tracking. This research explores OpenCV techniques to build an intuitive human-computer interaction system controlled entirely through webcam-captured hand gestures.

OpenCV Computer Vision Hand Tracking HCI

Experience

AI/ML Computational Science Analyst

Accenture • Nov 2023 – Aug 2025 • Hyderabad, India

Built scalable AI features for the GenLite reverse engineering platform using LLMs, function calling, and advanced prompt engineering to enable seamless cross-language code transformation. Designed and implemented a code conversion module with integrated preprocessing pipelines, reducing manual code migration effort by over 70%. Delivered multiple POC projects for RAG applications and AI chatbots, with several promoted to full-scale development and production deployment.

LLMs RAG Azure Production AI

Project Intern

Samsung PRISM • Sep 2021 – Apr 2022

Researched model interpretability with a team of 6 using SHAP, CEM, and LIME, achieving a 40% improvement in explainability. Developed an explainable AI (XAI) framework for production ML systems to reduce reliance on black-box models and improve interpretability.

Explainable AI SHAP LIME Research

Education

Master of Science in Engineering Data Science and AI

University of Houston • Aug 2025 - May 2027

CGPA: 4.0/4.0

Data Science Artificial Intelligence Machine Learning

B.Tech in Computer Science and Engineering (Artificial Intelligence)

Amrita School of Engineering, Bengaluru • June 2019 - May 2023

CGPA: 8.43/10

Computer Science AI Engineering Deep Learning

What's Next?

I'm actively seeking new opportunities and collaborations where I can contribute to innovative, data-driven products.

My inbox is always open — whether you have a question or just want to say hi, I'm happy to chat about AI, tech, or anything in between!

Get In Touch