Amazon Lex is a deep learning service provided by AWS that powers conversational bots, which can be text- or voice-activated. It's used to perform actions, such as ordering a pizza, and is referred to as intents. Each bot may perform multiple intents, each triggered by distinct phrases called utterances. The Natural Language Understanding (NLU) power of Lex bots shines in its ability to define sample utterances and infer all possible variations. A fictitious movie search scenario was used to illustrate how Lex works, where a user's utterance is matched against predefined sample utterances to trigger the "SearchMovies" intent. If the user's utterance veers too far away from the sample utterances, Lex would stop detecting the match. To fulfill the intent, additional information such as movie genre and release year is required, which are called slots in the Lex terminology. Slots are collected one at a time after a specific Lex prompt. A demo bot scenario was created to search for movies among those stored in a MongoDB database running in MongoDB Atlas, using the same fictitious example. The data setup and exploration involved deploying the movie database to MongoDB Atlas and examining the structure of a typical document. Key properties of interest were identified as principal cast members, genres, and release year, which will be leveraged in the Lex bot and Lambda function. Secondary indexes on these fields are recommended for faster queries.