-
Jun 18, 2024 | 
dev.to | Jonathan Ellis
In November, Cohere released a dataset containing all of Wikipedia, chunked and embedded to vectors with their multilingual-v3 model. Computing this many embeddings yourself would cost in the neighborhood of $5000, so the public release of this dataset makes creating a semantic, vector-based index of Wikipedia practical for an individual for the first time. Here’s what we’re building: You can try searching the completed index on a public demo instance here.
-
Apr 1, 2024 | 
thenewstack.io | Jonathan Ellis |Loraine Lawson |Joab Jackson |Kimberley Mok
Vector indexes are the hottest topic in databases because approximate nearest neighbor (ANN) vector search puts the “R” in RAG (retrieval-augmented generation). “Nearest neighbor” for text embedding models is almost always measured with angular distance — for instance, the cosine between two vectors.
-
Mar 17, 2024 | 
thenewstack.io | Jack Wallen |Gorkem Ercan |Manish Kapur |Jonathan Ellis
Python is an object-oriented programming (OOP) language, which means it is centered around the idea of “objects” that can contain data (such as attributes) and code (such as methods). Because Python is an OOP language, it allows for the creation of classes. What are classes? I’m glad you asked. A class is a blueprint that is used to help organize and structure code for reusability and modularity. A good way to think of a class is that it is similar to a blueprint for a house.
-
Mar 16, 2024 | 
thenewstack.io | David Eastman |Gorkem Ercan |Manish Kapur |Jonathan Ellis
An important career pivot for any developer — whether forced by layoffs or perhaps a desire to start something new — is moving from one industry sector to another. This might even involve a complete change — from gaming to medical instruments, banking to food manufacturing, etc. Regardless, to be attractive to a potential employer, you must not only bring fresh insights but also an encapsulated understanding of the sector you just left.
-
Mar 6, 2024 | 
thenewstack.io | Jonathan Ellis |Tim Davis |Yujian Tang |Gorkem Ercan
Retrieval augmented generation (RAG) is by now a standard part of generative artificial intelligence (AI) applications. Supplementing your application prompt with relevant context retrieved from a vector database can dramatically increase accuracy and reduce hallucinations. This means that increasing relevance in vector search results has a direct correlation to the quality of your RAG application.
-
Feb 29, 2024 | 
thenewstack.io | Rak Siva |Gorkem Ercan |Manish Kapur |Jonathan Ellis
AI copilots are making significant strides in the coding landscape, providing real-time assistance, automating routine tasks and offering insightful suggestions from vast code repositories. These copilots deliver more than generic advice; they provide contextually relevant suggestions and generate code snippets as you type, akin to having a seasoned developer guiding you. However, we’re still evolving in the realm of code generation.
-
Feb 28, 2024 | 
thenewstack.io | James Walker |Gorkem Ercan |Manish Kapur |Jonathan Ellis
A data platform is how a company stores, manages and analyzes its data, its most valuable asset. The more powerful and efficient the platform, the more effectively the data can be put to use. Data platforms enable conversion of data streams from sources of all kinds, from line-of-business apps to IoT platforms to AI tools, into plans of action for achieving business outcomes like product improvements, better processes and new commercial opportunities.
-
Feb 28, 2024 | 
thenewstack.io | Anirudh Ramanathan |Gorkem Ercan |Manish Kapur |Jonathan Ellis
When doing end-to-end or acceptance testing, there are many ways to allow developers to experiment in well-isolated environments. Asynchronous architectures present a different set of challenges without a single set of best practices for replication and testing, which makes implementing a testing solution more complex.
-
Oct 9, 2023 | 
dzone.com | Jonathan Ellis
Vector search is a critical component of generative AI tooling because of how retrieval augmented generation (RAG) like FLARE helps LLMs incorporate up-to-date, customized information while avoiding hallucinations. At the same time, vector search is a feature, not a product — you need to query vectors as they relate to the rest of your data, not in isolation, and you shouldn’t need to build a pipeline to sync the rest of your data with a vector store to do that.
-
Sep 22, 2023 | 
dzone.com | Jonathan Ellis
With the huge demand for vector search functionality that’s required to enable generative AI applications, DataStax set an extremely ambitious goal to add this capability to Apache Cassandra and Astra DB, our managed service built on Cassandra. Back in April, when I asked our chief vice president of product officer who was going to build it, he said, “Why don’t you do it?” With two other engineers, I set out to deliver a new vector search implementation on June 7 — in just six weeks.
