Vector search is a method that uses machine learning AI models to represent semantic concepts with numbers, enabling the comparison of records. It has become crucial for ecommerce sites due to its ability to handle ambiguous language and provide more accurate results than traditional keyword searches. The technique involves encoding linguistic meaning into vectors, which can then be used for mathematical operations such as addition, subtraction, and similarity calculations. Vector search is not only simple to create but also continually improving thanks to advancements in machine learning technology. It has proven effective in connecting customers with relevant products, leading to increased conversion rates.