Software & Data Engineer building: 🌎 https://nitter.cf/t.co/mXsKJ361fq 👩💻 https://nitter.cf/t.co/ZpkD4tYqkG
Joined July 2007
- Tweets206
- Following435
- Followers115
- Likes226
Maurício Maia retweeted
I tried to cover the fundamentals of vector search and how to make search queries more efficient using @postgresql and @pgvector.
Is there anything else I should have added?
neon.tech/blog/understanding…
Maurício Maia retweeted
1/ 🤖 You don’t need a specialized vector database. You just need PostgreSQL.
Introducing Timescale Vector: PostgreSQL++ for AI Applications
- 📈 3x ANN search performance vs Weaviate, 40%-1,500% boost for pgvector.
- ⚙️ No need to learn and manage a separate vector database.
Interesting new position spotted at @sourcegraph: LLM Interaction Engineer
aijobnetwork.com/jobs/source…
#jobs #remotejobs
105
There are 9 open roles in the new OpenAI's Dublin office
We’re opening an office in Dublin, Ireland 🇮🇪 openai.com/blog/introducing-…
🗂️ Data from AIJobNetwork.com where I track jobs from selected AI companies.
Analysis inspired by the amazing work of @AznWeng
64
Excellent analysis of the market for AI talent at big companies.
Companies are hiring like crazy for Generative AI talent. In this 🧵 I’ll dive into:
- Companies leading the hiring surge (+1 that stands out)
- Technologies/cloud services most mentioned in these jobs
- How much $$$ companies are paying for GenAI talent
- Common use cases
1
1
239
Maurício Maia retweeted
Jonathan Katz (@jkatz05): pgvector 0.5.0 Feature Highlights and HOWTOs postgr.es/p/68a
Maurício Maia retweeted
Here goes: viberary.pizza is a side project I've been working on over the past few months. It gives you book recommendations based on ✨vibes✨.
You enter a search query like "funny scifi" and it returns a list of good recommendations (still working on the good part).
Maurício Maia retweeted
Today, we are taking your AI and LLM apps to the next level🚀
Introducing pg_embedding. A new extension for @Postgres and @langchain for scalable AI apps.
Learn how #pg_embedding outperforms #pgvector by 20x in speed with HNSW index 🎉⬇️
bit.ly/46HykGF
At some point, anyone building production-grade apps will realize this.
Understand your data and optimize splitting it.
(6/14) The first attempt was to naively split the custom knowledge base into equal-length chunks. The idea was to pass relevant excerpts to LLM. Good in theory, but it led to mismatched advice and objections. We needed to maintain semantic boundaries.
2
127
Maurício Maia retweeted
GPT-4 and GPT-3.5 Turbo models in the API now support calling your custom functions, allowing the model to use tools you design for it. Also — reduced pricing & new model versions (including 16k context for 3.5 Turbo): openai.com/blog/function-cal…
AIJobNetwork is on Product Hunt today producthunt.com/posts/the-ai…
It's more of an exercise to put something out there than anything else. #buildinpublic
3
151
Today on Product Hunt, there are three exciting launches for those who are working with OpenAI 👇
1
1
154
LLM Report producthunt.com/posts/openai…
1
1
90