Build high-performance RAG using just PostgreSQL (Full Tutorial)

🔗 GitHub Repositories

🔗 Try pgai on Timescale free for 30 days

⏱️ Timestamps
00:00 Introduction
02:10 Why PostrgeSQL + Pgvectorscale?
06:04 Set up Docker environment
10:23 Vectorizing and inserting the data
20:02 Perform vector and filtered search

👋🏻 About Me
Hi! I'm Dave, a Data Scientist turned AI Engineer and founder of Datalumina®. On this channel, I share practical tutorials to help you become better at building data + AI applications. If you want to know how I help data professionals beyond these videos, then check out the links on my channel.

#rag #timescale #pgvectorscale

[jeelanshahtlyr]

Dave, even though I'm not a developer, I find myself captivated by your content. Your detailed, step-by-step explanations make me feel like I'm part of a classroom full of developers. It's truly impressive how immersed you are in explaining the code. Even though half of it goes over my head, I still find it incredibly engaging and informative. Thank you for making complex topics so accessible!

[Sergio-rqmm]

This couldnt have come at a better time for me. This is exactly what I needed right now. Such a great video/tutorial. Thank you so much for this and for sharing your code.

[ilyanemihin]

Thanks! For me, this is the best practical explanation of RAG. I will definitely use this approach in my projects.

[TimescaleDB]

PostgreSQL is all you need! Great video, Dave!

[uxaiagency]

Congrats on 100k. Another great video, creating quality rag systems is what I was learning about and this is very clarifying

[Poosboy]

im sure you've streamlined this even more over time, but this was very helpful thanks

[krieger]

Hi Dave, great video for introducing into the capabilities of postgres with pgvector, Thanks a lot.

[paraconscious]

you are an Angel Dave!! God bless! This is immense value addition, thanks!

[sjkba]

It's the first time I'm seeing one of your videos, and I love it. Great presentation and content. Kudos for providing the Github link.

[sma]

Love your approach. Explain things for better understanding + Practical hands on tutorial. <3

[rembautimes]

I really respect Timescales engineering skill in enhancing PG 😊

[motivationgoat]

Congrats on 100k subs, awesome tutorial

[AlfianFirmansyah]

This is what exactly I needed, I have been trying for many tools like milvus, mindsDB, however, the integration with postgres is somehow not as smooth as timescale/vectorscale. greatly appreciate it, and thx for code snippet!

[DavidOden-kknm]

This is amazing man, thanks for this 😄

[VisibleProjects]

This is amazing - thank you so much 🙌🏼

[nufh]

Thanks, you give me some idea on how to work on my project.

[Brunocam]

Nice vid! Could you do a more advanced one? Like making the llm decide which metadata to filter or graph rag?

[luciano]

Hi Dave, thanks for a great tutorial! I ran into a compiler issue for psycopg2 (needed for timescale-vector). I could solve this by installing psycopg2 from conda using command: conda install -c conda-forge psycopg2. This worked for me instead of using pip.
Perhaps this is useful for someone who ran into the same issue...

[testadrome]

Great video Dave! Can you explain how to implement hybrid search on top of this i.e. integrate this together with a keyword-based search?

[orlandocastellanos]

Grate video, cna you make a full tutorial building an Ai agent work flow?