## Centralized Feature Management

This project shows how to use Feast to manage features consistently between model training (offline) and real-time inference (online). The goal was to build something small but production-like, so I set up a real feature store pipeline instead of just using Pandas in memory.

## Semantic Search

This project involves the development of an advanced semantic search system tailored for real estate listings. Leveraging the latest updates in Elasticsearch and integrating LLM to JSON, I aim to significantly enhance search accuracy and relevance.

## Word2Vec

Implementation of the first paper on word2vec [Efficient Estimation of Word Representations in Vector Space](https://arxiv.org/abs/1301.3781)Efficient Estimation of Word Representations in Vector Space. While there are 2 proposed model architecures in the paper (cbow & skipgram) I focused on the Continuous Bag-of-Words Model (CBOW), that predicts a words based on it's context.