Inline Code Code is now supported in Amazon Bedrock Glows on Social Test

nimda August 21, 2025

0 1 5 minutes read

Inline Code Code is now supported in Amazon Bedrock Glows on Social Test

Today, we are very happy to announce the public view of Inline Code for Amazon Bedrock Glows. Fig-powerful force, you can write Python texts directly within your work travel, reducing the need for different Wasmbolder's unique LOGIC Works. This feature distributes the first power and potrocessing fun (such as the usual data and feedback), to facilitate the development of AI app and make it more accessible to organizations. By removing steps of steps and reducing the maintenance overhead, the Inline Code is early in Enterprise acceptance of AI-generating AI, which leads to the speedy contouration of the AI application.

Organizations that use Amazon is now using Inline Code design and sending applications for building AI united AI requests and active in Amazon Bedrock while accessing the following:

Reorganization – Converting Input Data Before submitting it to the larger Language (LLM) model without setting a different lambda function. For example, emitting certain fields from JSON, formatting text data, or standard prices.
Posting back – Making tasks on the removal of models directly within the flow. For example, uninstalling businesses from the answers, formatting JSON about the Downsmeam, or using business rules in effects.
Cases used by complexity – Managing the performance of difficult traveling areas, which work with a AI can call famous packages such as OpenCV, Sciph, BYPDF.
Builder-Friendly – Creating and managing Inline Code through Amazon Bedrock API and AWS Management Console.
Visualization – Sexperediented user experience with the ability to follow input and out of each area.

In this regard, we discuss the benefits of this new feature, and show how we can use Inline Code in Amazon Bedrock Glows.

Benefits of Inline Code in Amazon Bedrock Flows

Thomson Reuters, International Information Services Company provides important issues, details, and solutions, financial, accompanial charges, charges of using many AIs need to be pre-operational. In an Inline Code of Amazon Bedrock Flows, Thomson Reuters can now benefit from the following:

Management has been reset – Transmit the need to create and maintain individual lambda activities with each custom block
The adjustment of the variable data – Enable Directorate Data directory before ILM calls and PostproSing of LLM OLM answers, including the ability to communicate internal AWS and APIs of the third time.
DIY flow creativity – Help users create difficult flow of code by Code by customization by using the interfering interface, without disclosure to infrastructure or need Lambda function management.

Looking for everything

In the following sections, we show how you can build a simple Amazon Bedrock's bed flow and add Inline code codes. Our Example Displays an effective app where we will build the flow processing user applications for Music Playlists, including both methods of using follow-up items and responsiveness.

Requirements

Before using new skills, make sure you have the following:

After these items, you can continue using Amazon Bedrock Flows with Inline Code codes in your productive AI system.

Create your flow using Inline code codes

Complete the following steps to create your flow:

In Amazon Bedrock Console, Select Flow behind Tools for Building to the wavering pane.
Create a new flow, for example, easy code flows. With detailed instruction commands, see the Amazon bed flow is now available for advanced safety and tracking.
Enter the Inline Code code. (As a result of this example, we built two different elevations).

Amazon Bedrock provides various types of node to build your fastest movement. For example, we use the code inline instead of calling a lambda function with a custom AI functionality code. There are two areas of the inline code in the transit. We are spreading from the documents create one faster flow. New kind of Node Inline code in Nodes tab on the left window.

Enter a specific code to process Preprocessing_InlineCode Node before sending it to a quick place prompt_1. Python 3 is only supported at the time of writing. In this example, we look at the number of songs requested by a user more than 10 and set up to 10.

There is a Python code editor and sample code templates and writing code.

We use the following code:

import json
def __func():
    try:
        if userprompt['number'] > 10:
            userprompt['number']=10
            return userprompt
        else:
            return userprompt
            
    except Exception as e:
        return {
            "error": "Invalid input format",
            "details": str(e)
        }
__func()

In the postrocesing_inline code code prompt_2.

def __func():
    # Remove extra whitespace and count
    cleaned_text=" ".join(playlist.split())
    word_count = len(cleaned_text.split())
    return{
        "playlist": playlist,     "word_count": word_count
    }
__func()

Check the quick flow of the following:

Sample input for the Flow Input node 
{
  "genre": "pop",
    "number": 8
  }

Input inline and long (Python function) must be treated as non-reliable users of the user, as well as appropriate insertion, verification, and data management must be used.

You can see the result as shown in the following screenshot. The program also provides access to NONE traces of NODO, providing detailed information in each process of processing, actual time operators, and highlight any issues that occur during the flow. Femons may be empowered using API and sent to Amazon Cloudwatch Log. At API, set enableTrace the true field in the InvokeFlow Ask. Each flowOutputEvent In response to return aside a flowTraceEvent.

You have now successfully created and made Amazon Bedrock flowing using Inline code codes. You can also use Amazon Bedrock Apis for orderly to make this flow. For more details how you can set up advanced safety and tracking, see the Amazon Bedrock flow is now available in advanced safety and tracking.

Consideration

When working with Inline Code Reference in Amazon Bedrock flow, the following are important things to be careful:

Code is made on the AWS, secure, the nature of the sandbox that can be charged with anyone and has no Internet access
The feature supports Python 3.12 and above
Treats the code in the size of binary size up to 4 MB, equivalent to 4 characters five
Sponsoring popular packages like OpenCV, Scipy, and Pypdf
Sponsoring 25 Same of AWS Code kill

Store

The integration of the Inline Code of Inline Code Bedrock Important Development Code in democracy, reducing the severity of different lambs of basic processing activities. This is enhanced directly to the Enterprise customer needs to receive more developmental activities, to help develop developers to focus on creating a complex AI system rather than controlling infrastructure.

Amazon Bedrock Bedrock Code is now available on the Social View of the following AWS locations: US East (NO. To learn more, see Creating your first flow in Amazon Bedrock and track each step with your flow by watching their trend at Amazon Bedrock.

We are happy to see new apps to create with these new skills. As regularly, we accept your feedback using AWS Re: Send Amazon Bedrock or contacts of the usual AWS. Join the building of AI's building community.Aws to share your information and learn from others.

About the authors

Shubankar Sumar Is the construction of senior solutions in AWs, where he works in creating the solutions of A-Power produced business software and Saas companies throughout the UK. Through the solid domain in the software engineering, Excerels in creating safe, false programs, and the cheapest multi-tenant Tenant in the cloud. His expertise is to combine the power of a cut-edge of AI for existing SAAS applications, to help customers stay before the new technologies.

The jesses are soi High product manager in Amazon Bedrock, ASS DENTIONA DEEDIFIZA device. She works in the communication of AI and the interaction of the Building and Improving AI products and services to produce our needs. Earlier, Jesse held the team engineering team of the party in Apples and Lumeds, and he was a great scientist in Licon Valley's implementation. You have MS and Ph.D. From the University of Florida, along with MBA from the University of California, Berkeley, a business school.