Welcome to my GitHub portfolio! I'm Rohith Ramakrishnan, a passionate and results-driven AI and Data Science professional with a strong academic foundation and hands-on experience in developing and deploying cutting-edge solutions. This repository showcases a collection of my projects, contributions, and publications.
I hold a B.Tech in Artificial Intelligence from Amrita Vishwa Vidyapeetham with a CGPA of 9.3/10 and I'm currently pursuing a Diploma in Data Science and Programming from IIT Madras. My expertise lies in leveraging advanced techniques in machine learning, natural language processing, and big data to solve complex real-world problems.
- BM25-Fusion Python Library: Developed a Python library, BM25-Fusion, for optimized big data indexing using numba and jit for sparse indexing.
- Llama-Index Integrations: Implemented streaming and chat support for NVIDIA's Triton inference in Llama-Index and S-BERT Re-Ranker Inference for Hugging Face's Text-Embedding-Inference in Llama-Index[cite: 3, 4].
I'm an active contributor to the open-source community, with notable contributions to:
- Llama-Index
- Hugging Face
- NVIDIA Triton Inference Server
I'm also a published author in the field of AI and Data Science:
- PACLIC-35: Analysis of Text-Semantics via Efficient Word Embedding Using Variational Mode Decomposition
- ISCMM-21: CT Image Enhancement using Variational Mode Decomposition for AI Enabled COVID Classification
- ICICC-22: Data-Driven Volatile Cryptocurrency Price Forecasting via Variational Mode Decomposition and BI-LSTM
Frameworks: Llama-Index, Langchain, Triton, TensorRT-LLM, SpaCy, VLLM, TensorFlow, PyTorch, Flask, Spacy, Dask, Docker, PySpark, Apache SOLR or Led Innovative Developments in GenAI LLMs [cite: 4]
Thank you for visiting my GitHub profile! Feel free to explore my repositories and get in touch if you have any questions or collaboration opportunities.