Hello everyone š Ahmed here. Iām the author of The Tech Buffet, a newsletter that demystifies complex topics in programming and ML from an engineering perspective.
Consider subscribing to access exclusive content.
In this issue, I list the previous articles I wrote on Retrieval Augmented Generation.
If youāre new to The Tech Buffet, this might be useful to catch up on the content you missed: explaining RAG mechanisms and unraveling their pitfalls, applying RAGs to SQL generation, getting RAGs to work well in production, different indexing and retrieval methods, etc.
These posts provide my perspective. After spending several months deploying RAGs in production for my clients, I came to know their ins and outs.
Expect this list to be regularly updated, serving as an evolving glossary for easy reference.
The Tech Buffet #1: How To Design a System To Chat With Your Private Data
Step-by-step tutorial to understand the high-level picture behind RAG: what components are involved and how do they interact with each other?
The Tech Buffet #2: How To Use LangChain to Perform Question Answering Over Documents
How to use LangChain to build a simple RAG that connects the different pieces: data loaders, text splitters, embedding model, vector store, LLM.
The Tech Buffet #4: Turn Complex English Instructions Into Executable SQL With LLMs
RAGs are typically used to answer questions over documents. In this issue, we explore another application that RAGs handle particularly well too: generating SQL queries from plain English instructions.
The Tech Buffet #6: Why Your RAG is Not Reliable in Production
Weāve all seen those demos of chatbots conversing with PDFs or emails.
While these systems are certainly impressive, they might not be reliable in production without tweaking and experimentation.
In this issue, we explore the problems behind the RAG framework and go over some tips to improve its performance.
The Tech Buffet #12: Improve RAG Pipelines With These 3 Indexing Methods
In RAGs, the documents you retrieve are not necessarily the documents you index.
For example, you can index a paragraph by the question it answers.
Discover how you can index data differently to meet your needs.
The Tech Buffet #13: Getting a RAG To Work Well Is Hard - 5 Blog Posts To Become a RAG Master
The RAG literature is flourishing. In this issue, discover 5 industry-driven blog posts that explain how RAGs are implemented in companies to scale and solve business needs.
The Tech Buffet #16: Quickly Evaluate your RAG Without Manually Labeling Test Data
Evaluating a RAG-based system is hard: this requires manually labeling a test set, some human expertise, and computing complex metrics.
This issue presents a solution to evaluate RAGs without any human intervention.
The Tech Buffet #17: 9 Effective Techniques To Boost Retrieval Augmented Generation (RAG) Systems
In this issue, we explore this paper: Retrieval-Augmented Generation for Large Language Models
It covers everything you need to know about the RAG framework and its limitations. It also lists modern techniques to boost its performance in retrieval, augmentation, and generation.
The Tech Buffet #18: Advanced Retrieval Techniques for RAG
The documents you retrieve from the vector store are not always relevant to the intention behind the user's query. They might also be redundant or missing crucial information.
To improve document retrieval, we explore 3 techniques: query expansion, cross-encoder re-ranking, and embedding adaptors.
I hope youāll find something that interests you in this list. If you have any questions regarding one of the issues, donāt hesitate to send me a message or drop a comment here. If you also have another topic in mind about RAGs, let me know.
In future issues, The Tech Buffet will cover practical tutorials on programming and DevOps. The goal is to help you build ML solutions (like RAGs) that scale in a production environment.
Thereās already a backlog of posts, but if youāre interested in a particular topic, Iād love to hear your suggestions.
Until next time š
Ahmed