A RAG application to search through the recent ML research without going through the Papers with Code manually.

CustomerCareBot is a Streamlit web application designed to automate customer service responses using AI-powered language models.

Example to help understand how to use hugging face sentance_transformers to encode searchable text into embeddings AND how to use cosine similarity search to determine similarity between a search prompt and the embeddings

Data structures for Similarity Search (Metric Space)

A simple npm utility module for obtaining similar commands in Discord Bots projects.

🏋️ Evaluate sentence similarities with standard metrics for NLP related tasks

This is a demo of how you can build an ml pipeline with isolated dockerized Jina microservices

Python implementation of the Vantage Point Tree index.

ML project to simulate a online fashion retailer - showcasing ranking of search results, recommendations using text and visual features

A snakemake workflow for iterative sequence profile searching using jackhmmer from the HMMer3 package.

Extending the LIDIA non-local denoiser to (i) videos and (ii) standard resolutions

Use NLP and clustering on movie plot summaries from IMDb and Wikipedia to quantify movie similarity

Creating a chat application that can handle multiple files using LlamaIndex, OpenAI, and Streamlit involves several steps

Legal case retrieval challenge. Solution based on similarity search and learning-to-rank methods

🔍 Work Wiz 🌟 Work Wiz is a job search platform that matches users with jobs based on their unique skillsets and desired pay, supporting both full-time and freelance opportunities. Using advanced technologies like TF-IDF, cosine similarity, and SVD from Sklearn, we ensure highly relevant search results.

Blog posts for ParadeDB as .mdx, hosted on Mintlify

Entity matching on the DBLP-ACM dataset