Skip to content
View Keerthanajan08's full-sized avatar

Block or report Keerthanajan08

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Keerthanajan08/README.md

Hi, I'm Keerthana Sasidaran 👋

AI/ML Engineer • NLP & Multimodal Systems • M.Tech, CSE

Building at the intersection of Machine Learning, NLP, and Large Language Models — from fine-tuning LLaMA 2 for hate speech detection to automating validation pipelines at Philips Innovation Campus.


🔬 About Me

  • 🎓 M.Tech in Computer Science, MSRIT — CGPA 8.57, First Class with Distinction
  • 🎓 B.Tech in Computer Science, MSRUAS — CGPA 8.88/10
  • 🧠 Research focus: Multimodal AI, NLP, Large Language Models
  • 📄 Published in IEEE Xplore (2024) and an IGI Global book chapter
  • 🏆 1st Place, IGT Bangalore Hackathon
  • 💼 Former Software Engineering Intern @ Philips Innovation Campus
  • 📚 Currently pursuing the AWS AI & ML Scholars program

🚀 Featured Research

Multimodal Hate Speech Detection using Fine-Tuned LLaMA 2 IEEE Xplore, 2024

  • Built an end-to-end multimodal pipeline processing text, audio, and video
  • Fine-tuned LLaMA 2, integrated Whisper and the Google Speech Recognition API for audio transcription
  • Benchmarked against BERT, RoBERTa, DistilBERT, ELECTRA, and XLNet
  • Achieved 97% accuracy and a 0.96 F1 score

📖 Also published: "Machine Learning for Big Data" — book chapter in Applications of Big Data in Large- and Small-Scale Systems (IGI Global)


🛠️ Tech Stack

Languages Python SQL

AI / Machine Learning

TensorFlow Keras PyTorch Hugging%20Face Transformers Pandas NumPy Scikit-Learn

NLP & Generative AI

Llama%202 Whisper Google%20Speech%20API

Backend & APIs

REST%20API Postman

Tools

Git GitHub Hyper-V Windows%20PE


💼 Professional Experience

Software Engineering Intern — Philips Innovation Campus · 11 months

  • Performed functional, integration, and automation testing across the software lifecycle
  • Automated validation workflows with Postman and PowerShell, meaningfully cutting manual validation effort
  • Debugged memory consumption, platform stability, and installation defects using memory profiling and log analysis
  • Built PoC environments with Hyper-V and Windows PE to validate installation workflows and catch defects early
  • Developed an ML-driven predictive maintenance system to reduce downtime and support early fault detection

🏆 Hackathons

IGT Bangalore Hackathon — 1st Place Built a remote diagnostic solution using PXE Boot and Windows PE.


📜 Certifications

  • Deep Learning Specialization — DeepLearning.AI
  • Advanced Deep Learning Specialist — IBM Digital Credential
  • Machine Learning with Python — IBM
  • Introduction to Deep Learning & Neural Networks with Keras — IBM
  • Deep Learning with Keras and TensorFlow — IBM
  • AWS AI & ML Scholars — in progress

📫 Let's Connect

Open to opportunities in AI/ML Engineering, NLP, and Applied Research.

Pinned Loading

  1. MultiLingualHateSpeech MultiLingualHateSpeech Public

    An end-to-end AI content moderation platform that analyzes text, audio, and video to detect hate speech using Large Language Models and multimodal AI.

    Jupyter Notebook

  2. Movie-Recommendation-System-with-Sentiment-Analysis-master Movie-Recommendation-System-with-Sentiment-Analysis-master Public

    Built an intelligent movie recommendation system leveraging content-based filtering, NLP techniques, and cosine similarity for personalized recommendations.

    HTML

  3. LLM-Prompt-Security-Gateway LLM-Prompt-Security-Gateway Public

    FastAPI gateway that detects pattern-based and semantic leakage of sensitive data in outbound LLM prompts, using regex rules, embedding-similarity retrieval against a confidential corpus, LangChain…

    Python