Hello everyone,
The highlight of this release is the introduction of generative AI in the Software Factory. We have added four Large Language Model connectors that you can use in your system flows. This opens a world of new possibilities.
Some other important features are the improved merging process, the introduction of public API roles, and the possibility of encrypting key values in a runtime configuration. From IAM, you can now send email notifications to users.
In this summary, you will read about some of the key features of this release. Read all the detailed release notes in the Thinkwise Documentation.
Contents
Highlights of this release
Generative AI in process connectors
As mentioned in the introduction, generative AI is now available in the Software Factory. To support you in using generative AI in your applications, we have added four Large Language Model connectors that you can use in your system flows.
Generative AI is a class of artificial intelligence techniques and models that is designed to create or generate new content, such as text, images, music, and more, by learning patterns and structures from large datasets.
A Large Language Mode (LLM) is an example of Generative AI. LLMs are large neural networks that are trained to predict the next token (word, punctuation, etc.) given some text. They are trained on very large amounts of data. You can instruct an LLM on what to do with your text.
The new connectors open a world of new possibilities. You can use them, for example, to:
-
Generate text based on a few keywords
-
Summarize text
-
Classify text (extract the subject or tags from a text)
-
Analyze the sentiment of a text
-
Answer questions
- Translate text to another language
The new connectors are:
- LLM Chat completion - use this connector to build a chat conversation between a user and a Large Language Model (LLM).
-
LLM Completion - this connector can suggest how to complete a provided text. You can use it, for example, to complete product descriptions, draft emails, or generate code.
-
LLM instruction - this connector can transform text following your instruction. It can, for example, summarize, translate, change the formality of, or remove sentiment from a text.
-
LLM embedding - this connector can find semantically related texts.
Share your use cases and inspire others in the Thinkwise Community!
Generative AI in the Thinkwise Platform
Using the new LLM connectors, we have introduced AI in the Software Factory and the Upcycler.
(available soon) In the Software Factory, you can use AI to perform a code review on any control procedure automatically. This feature is intended to supplement human reviewers and should be used to assist in the review process. You can still deactivate the AI review if you prefer a manual review.
In the Upcycler, AI is now available to enhance the enrichment capabilities. Upcycled applications typically contain code that is written in a different programming language, is missing comments, or contains comments in another language. With Ai integration, you can now use OpenAI to, for example, add comments to existing code, add descriptions for any object in the model, or replace existing code concepts with new ones.
To use these features, you must enable AI for the Software Factory or Upcycler.
Branching and merging
We changed the detail screens of the merge process to improve the process and the conflicts overview.
- Tab Conflicts is now called Active conflicts. It contains a list of all conflicts that have been detected after you started the merge session. On the new detail tab Compare [object type], you can quickly identify and focus on specific conflicts. It also provides historical context so you can resolve conflicts more efficiently.
- Tab Actions is now called Advanced. It contains a complete overview of all generated delta actions. For more complex merge sessions, you can select here if and how actions should be applied or ignored.
- Tab Merge impact is new. It contains an overview of all the actions that will be done during the execution of the merge session. You can use this for a full review.
In addition, merge sessions now also create and automatically resolve conflicts for duplicate update and delete actions in the source and target branches. This was already available for duplicate insert actions.
Public API roles
To integrate your application with other systems, you can use Indicium calls with authentication, for example, as a client application or with an OpenID provider.
However, systems like webhooks or subscriptions do not always allow authentication.
In these cases, you can now make the API call public for access without authentication. You can use this for actions like processing appointments and emails with Microsoft Graph.
To set up a public API call in the Software Factory, you must assign a role to the object, such as a subroutine, and make it public with the new checkbox Allow as public API. The next step is activating or deactivating a public API role in IAM.
Other new and changed features
Email notifications from IAM
You can now send email notifications from IAM to users. Use this, for example, to inform users of an application about planned system downtime.
Encryption of key values
You can now encrypt key values in a runtime configuration for a File storage location, Email provider, Generative AI provider, or OAuth server.
In addition, we have improved the encryption process when you switch to another type of email provider or file storage location in the Software Factory or IAM. This improvement gives you more control over the use of encryption.
Questions or suggestions?
Questions or suggestions about the release notes? Let us know in the Thinkwise Community!