Skip to content

Latest commit

Β 

History

8 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 

Repository files navigation

GitHub Banner

Hi πŸ‘‹, I'm Devasish Viswanadh Kolla

AI Researcher β€’ Cybersecurity Enthusiast β€’ Full-Stack Developer

Portfolio β€’ LinkedIn β€’ Email


πŸš€ About Me

πŸŽ“ Final-Year B.Tech Computer Science Student @ SRM University AP

πŸ”¬ IEEE Published Author

πŸ“œ Patent Applicant in Trustworthy AI

☁️ AWS Certified Cloud Practitioner

πŸ›‘οΈ Cybersecurity Intern @ APTS & APCSOC (Govt. of Andhra Pradesh)

πŸ€– Passionate about AI, NLP, Explainable AI, Cybersecurity, Cloud Computing, and FinTech

🌱 Currently working on:

  • Loan Default Risk Analyzer using Explainable AI
  • AI Hallucination Detection Research
  • Secure Backend Systems with Spring Boot
  • Web Application Security Assessments

πŸ† Achievements

  • πŸ“„ IEEE Published Author
  • πŸ“œ Patent Application Filed
  • ☁️ AWS Certified Cloud Practitioner
  • πŸŽ“ Merit Scholarship (75% Tuition Waiver)
  • πŸ₯‰ Bronze Medalist – Karate Kata
  • πŸ›‘οΈ Cybersecurity Intern – Government of Andhra Pradesh

πŸ“š Research & Publications

πŸ“„ The Perils of Naive Truncation: A Context Ablation Study for Dialogue Summarization on DialogSum

  • Published at IEEE AISP 2025
  • Evaluated dialogue truncation strategies using transformer models
  • Analyzed effects of context ablation using ROUGE and BERTScore

πŸ“Œ Patent Work

System and Method for Detecting Hallucinated Information in AI-Generated Summaries Using Multi-Stage Verification Architecture

  • NLI-based verification
  • Composite Faithfulness Risk (CFR) Score
  • Multi-stage hallucination detection
  • Sentence-level fact validation

🌐 Connect With Me


πŸ’» Tech Stack

Java Python C++ SQL

PyTorch Transformers Scikit-Learn XGBoost SHAP LIME

Spring Boot FastAPI PostgreSQL MySQL

React TypeScript Tailwind CSS Vite

OWASP Burp Suite OWASP ZAP Wireshark

AWS Docker Git GitHub Actions Linux


πŸ“Š GitHub Statistics


πŸ”₯ GitHub Streak


πŸš€ What I've Been Building

πŸ€– Faithful Text Summarization

Explored the reliability of AI-generated summaries using BART, ROUGE, BERTScore, and NLI-based faithfulness evaluation. Focused on identifying hallucinations and improving trust in Generative AI systems.

πŸ“„ Dialogue Summarization Research

Author of the IEEE publication "The Perils of Naive Truncation: A Context Ablation Study for Dialogue Summarization on DialogSum". Investigated how context loss impacts summary quality in transformer-based dialogue summarization.

πŸ” Secure File Locker

Engineered a cryptography-driven file protection system in C++ using AES encryption, SHA hashing, and secure access controls to ensure confidentiality and integrity.

πŸ”‘ JWT Authentication Backend

Designed a secure Spring Boot backend implementing JWT-based authentication, authorization, and role-based access control for modern web applications.

πŸ’° Loan Default Risk Analyzer

Building an Explainable AI framework for NBFCs using ensemble learning, SHAP, and LIME to make credit risk predictions transparent and actionable.

🌐 Personal Portfolio

Crafted a modern, interactive portfolio showcasing research, projects, certifications, and technical expertise with React, TypeScript, Tailwind CSS, and Vite.


🎯 Areas I Love Exploring

🧠 Artificial Intelligence & Machine Learning

πŸ” Explainable AI (XAI)

πŸ“š Natural Language Processing

πŸ›‘οΈ Cybersecurity & Secure Software Development

☁️ Cloud Computing & Backend Systems

πŸ’Ή FinTech & Data-Driven Decision Making


πŸ’‘ My Mission

To build trustworthy AI systems, secure software, and scalable technology solutions that solve meaningful real-world problems while maintaining transparency, reliability, and user trust.

About

My GitHub profile README showcasing my projects, research publications, certifications, and technical expertise.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors