Making Data Simple · IBM Big Data & Analytics Hub

Databases That Think: Building AI-Ready Systems with MongoDB’s Richmond Alake {Replay}

October 8, 2025·41 min·3 clips

Richmond Alake, an AI practitioner with 200 technical articles and 1 million views, joins to discuss how MongoDB fits into the AI space.

The host wastes almost no runway. He thanks the Making Data Simple listeners, admits he is squeezed for time, and brings on Richman Alake with the casual ease of someone already calling him Rich. Rich has real technical mileage: computer vision, robotics, machine learning, software development, AI ML data classes, and more than 200 technical articles with 1 million views. The mood stays friendly, but the host keeps steering past the buzzwords toward the practical data story. Rich traces a winding path through maths, mechanics, software engineering, code, database systems, application development, and then AI as the next natural stop after web development. The examples give the conversation some weight. He talks about inverse kinematics, rover motion, deep learning, convolutional neural networks, and the older vector ideas sitting underneath today's AI language. Then the host presses the label. If MongoDB is going to be called an AI database, what does that mean? Rich keeps it plain. The better phrase is a database optimized for AI workloads, because MongoDB still stores and retrieves data like a database. That small correction matters. It moves the episode away from shiny language and into how AI apps actually use data. ChatGPT becomes the easy doorway. Rich explains that model answers come from parametric knowledge, while RAG adds relevant data to the prompt before the model responds. Retrieval Augmented Generation is not treated like magic. It is a way to give the model better context. Vector embeddings are the bridge, numerical representations of data with roots in computer vision and neural networks. The exchange stays loose and conversational, but the useful thread is implementation over slogans. The close circles back to prompting. The host says comparing and reconciling versions can help when the prompt is handled well. Rich leaves warmly, and the host thanks him for his contagious energy before inviting guest ideas at almartintalksdata at gmail.com.

As heard by us

A plainspoken chat on AI-ready databases, with a clear focus on RAG and vector embeddings.

Rich and the host keep things on the practical side, with most of the discussion centered on MongoDB as a database built for AI workloads, how RAG shapes retrieval, and why vector embeddings matter.

Read the full review in PlayNext →

Why you'd press play

You want a clear, no-nonsense take on how MongoDB fits into AI workflows.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Making Data Simple