Subscribe
Sign in
Home
Notes
Chat
DIY
Blog
Audio pills
Real-World
Extra
Archive
Leaderboard
About
RW #9 - How Vector DBs Store 100M Embeddings on One Machine (and Still Search Fast)
The claim “Our vector database can handle 100 million embeddings on a single machine.” Sounds impressive, but the math doesn’t work using just naive…
Jan 11
•
Nino Risteski
9
2
Issue #117 - Scaling Vector Search: HNSW and Approximate Search
With a million vectors, brute-force search (checking every single one) is like trying to find a specific person in a packed stadium by tapping every…
Jan 3
•
David Andrés
9
4
MLPills - Year in Review 2025
If we had to summarize 2025 in one sentence, it would be: The year we moved from “Prompts” to “Systems.”
Dec 28, 2025
•
David Andrés
13
2
Issue #116 - Introduction to Vector Search
For decades, search engines operated on a simple premise: exact matching. If you searched for “car troubleshooting,” the engine looked for documents…
Dec 21, 2025
•
David Andrés
6
2
Issue #115 - Reranking in your RAG pipeline
If you are building an AI application today, you are likely using RAG (Retrieval-Augmented Generation). It is the standard architecture for giving Large…
Dec 14, 2025
•
David Andrés
7
1
Most Popular
View all
Issue #88 - Introduction to SHAP values
Jan 29, 2025
•
David Andrés
and
Muhammad Anas
24
3
DIY #12 - SHAP in Action: Making ML Explainable
Feb 15, 2025
•
David Andrés
and
Muhammad Anas
19
2
2
Issue #67 - Exploratory Data Analysis
Jul 28, 2024
•
David Andrés
and
Josep Ferrer
24
2
RW #3 - EDA applied to Netflix (part I)
Mar 30, 2025
•
David Andrés
and
Muhammad Anas
16
2
DIY
View all
DIY #18 - Orchestrator-Worker LLM Agent with LangChain
Imagine you’re managing a complex research project. You need to analyze a company’s financial health, but that requires gathering quarterly reports…
Nov 23, 2025
•
David Andrés
9
1
DIY #17 - Parallelisation with LangChain
Imagine you’re a news editor trying to understand a breaking story. You get a single field report. To really cover it, you need to know the “who, what…
Nov 1, 2025
•
David Andrés
7
1
DIY #16 - Build a Persistent Conversational Agent with LangGraph
In this article, we'll build exactly that: a Python-based conversational agent using LangGraph that can remember user information across sessions and…
Sep 7, 2025
•
David Andrés
6
1
DIY #15 - Prompt Chaining with LangChain
Imagine you have a complex task, like writing a detailed report or analyzing a lengthy customer feedback document. Trying to get an AI to do it all in…
Jun 29, 2025
•
David Andrés
8
2
DIY #14 - Step-by-step implementation of a ReAct Agent in LangGraph
Large Language Model (LLM) agents can make decisions about when to use external tools as part of answering a question. Let’s now cover the most basic…
Apr 20, 2025
•
David Andrés
9
2
Real-World
View all
RW #9 - How Vector DBs Store 100M Embeddings on One Machine (and Still Search Fast)
The claim “Our vector database can handle 100 million embeddings on a single machine.” Sounds impressive, but the math doesn’t work using just naive…
Jan 11
•
Nino Risteski
9
2
RW #8 - Turning Podcasts Into Knowledge Graphs
Picture this: You’ve just finished listening to a fascinating 2-hour podcast. The expert dropped dozens of insights, research findings, and connections…
Sep 28, 2025
•
David Andrés
13
1
RW #7 - When to use Rules, ML or LLMs?
Deciding between a simple, rule-based system and a sophisticated machine learning (ML) model is a critical choice in software development. While it's…
Sep 20, 2025
•
David Andrés
14
4
RW #6 - Text-Moderation System with Embeddings
This week we’re handing you a plug-and-play, notebook-ready tutorial you can drop straight into Jupyter or VS Code. Inside you’ll find:Why an…
Aug 3, 2025
•
David Andrés
4
1
RW #5 - No-Code Customer Service agent with LangFlow
Imagine a customer service operation where AI agents don't just rely on generic responses, but actually understand the business inside and out. Agents…
Jul 13, 2025
•
David Andrés
7
2
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
