 |
VOOZH | about |
|
VOOZH | about |
We’re so glad you’re here. You can expect all the best TNS content to arrive Monday through Friday to keep you on top of the news and at the top of your game.
Check your inbox for a confirmation email where you can adjust your preferences and even join additional groups.
Follow TNS on your favorite social media networks.
Become a TNS follower on LinkedIn.
Check out the latest featured and trending stories while you wait for your first TNS newsletter.
PITTSBURGH — If you’re involved in machine learning and data science, you don’t have to use Python, but it certainly helps, according to Suman Debnath.
“When we start talking about machine learning, people tend to start with Python,” said Debnath, principal developer advocate for machine learning at Amazon Web Services, in this One the Road episode of The New Stack Makers, recorded at PyCon US in May.
“There are two reasons for that. One is it’s very easy to get started with. And second is, the whole ecosystem of data science and machine learning. It started with Python majorly, not with other languages.”
Building generative AI-based applications, however, doesn’t require the data science expertise that machine learning does. “The whole idea is how you can use these machine learning models for your application,” he said. “And for that, you can use any language of your choice.”
Including Python. In this episode of Makers, Debnath discussed how Python programmers can use Amazon Bedrock, AWS’s GenAI framework, to build AI-based apps.
The advantage of GenAI is that it can return results quickly in response to text prompts. Python, despite its advantages (such as being memory safe), is not known for its speed, Debnath told our Makers audience. (Though lots of people are working on that.)
Amazon Bedrock, introduced last September, is what Debnath called a “Model as a Service” framework. It’s “very easy for any developer to come on AWS and get added with it without knowing anything about machine learning, without knowing about what is happening under the hood with the different models,” he said. “All we are providing is an API-based service.”
He added, “You could be writing your code in Python, C, C++, Java, whatever it is, you can just use the API to use the goodness of all these large language models.”
Bedrock also accommodates retrieval augmented generation (RAG) so developers can customize the LLMs they’re using.
The language agnosticism is a benefit for Python users, Debnath said. “One good thing about Bedrock is it has a very good ecosystem with other third-party libraries, which are open source, like Langchain, llamaindex and all that. So you have the flexibility to use that within your existing framework. It’s not like you have to use Amazon’s SDK to get started with Bedrock.”
To get started with Amazon Bedrock, he recommends visiting the community.AWS platform and navigating to the “generative AI” space; the site includes information not only about Bedrock but also other GenAI-based services like Amazon Q. A workshop for Amazon Bedrock newbies is also available on GitHub.
Watch the entire episode to see Debnath demonstrate Amazon Bedrock and discuss its advantages for Python users.
