Passionate about building AI systems that combine Machine Learning, Computer Vision, and Large Language Models to solve real-world problems.

• Agentic RAG Pipelines
• AI Agents & Multi-Agent Systems
• LLM Orchestration
• Production-ready AI Applications

A six-stage multi-model fusion pipeline that generates, scores, and refines image captions using Florence-2-Large, BLIP ITM, Jina Reranker M0, Cosine Similarity, majority voting, and Qwen2.5-1.5B-Instruct — evaluated on the Flickr8k benchmark (8,091 images) and deployed as a live web app.

🔗 GitHub Repo · Live Demo

NLP-based classification system using TF-IDF vectorization and supervised ML models to detect fake news articles. Compared Logistic Regression, SVM, Naive Bayes, and Decision Tree across labeled datasets.

🔗 GitHub Repo

Analysis of how the Bitcoin Fear and Greed Index influences trader behavior and performance across 125,830 on-chain trades on Hyperliquid — covering win rate, position sizing, trade direction, and profitability across all five sentiment regimes with statistical significance testing.

🔗 GitHub Repo

I am a recent M.Sc. Data Science graduate from Aligarh Muslim University (AMU) with a strong interest in Machine Learning, Computer Vision, Deep Learning, and Generative AI.

My work focuses on developing intelligent systems that can understand and generate meaningful insights from data, images, and text. I enjoy building end-to-end AI solutions, from data collection and preprocessing to model development, evaluation, and deployment.

Recently, I worked on a multimodal Image Captioning Refinement System that combines Florence-2, BLIP, Qwen, Jina Reranker, and embedding-based evaluation techniques to generate high-quality image descriptions. I am also interested in LLM evaluation, Vision-Language Models, and AI-powered applications.

Currently, I am seeking opportunities as a Data Science Intern, AI Intern, or Entry-Level Data Scientist where I can contribute, learn, and grow while solving impactful problems.

• Deep Learning — Neural Networks (ANN, CNN, RNN, LSTM), Transformers, LLMs, Vision-Language Models (BLIP, Florence-2, Qwen)
• Computer Vision & NLP — Image Captioning, Multimodal AI, TF-IDF, Tokenization, Text Classification, Embedding Extraction
• Machine Learning — Logistic Regression, SVM, Naive Bayes, Decision Trees
• Data — Exploratory Data Analysis, Feature Engineering, Statistical Analysis, Predictive Modeling
• Visualization — Matplotlib, Seaborn, Tableau, Plotly
• APIs — Gemini API, Jina API

LLM Post-Training Intern — Ethara.ai Feb 2026 – Mar 2026

• Evaluated and annotated LLM outputs for quality, coherence, and alignment with predefined benchmarks to support model post-training.

Data Scientist Intern — Evoastra Ventures Inc. Oct 2025 – Nov 2025

• Built a deep learning–based image captioning system using CNN + LSTM
• Performed text preprocessing, tokenization, and vocabulary creation
• Conducted EDA and documented insights for real-world business use cases

• M.Sc. Data Science — Aligarh Muslim University (2024–2026)
• B.Sc. Mathematics — CCS University (2021–2024)

⭐ Feel free to explore my repositories and connect with me for collaboration or opportunities in Data Science & Machine Learning.