Quarterly Release Update: Introducing Tecton 1.1

We’re excited to share our latest release with Tecton 1.1. The Tecton team has been working hard building powerful capabilities that make it simpler for AI teams to build more sophisticated features, optimize infrastructure and improve model performance. Read on for a roundup of what’s new.

Seamlessly access any third-party data source in real time

Building effective features often requires accessing valuable external information, such as fraud signals, environmental data, and operational data. But orchestrating and managing these integrations is overwhelmingly complex.

Tecton’s API Resources simplify using third-party APIs. API Resources enable on-demand access to dynamic data from external endpoints, making it frictionless to create features with key third-party sources such as:

• Financial data (eg. Plaid) and fraud signals (eg. Socure, Sift)
• Environmental data (eg. OpenWeatherMap)
• AI Services: Leverage state-of-the-art models from OpenAI and AWS Bedrock to generate embeddings and other model-generated features
• Operational Data: Query mission-critical data directly from production databases like PostgreSQL for real-time feature computation

Walk through an example here to see how you can use an API Resource to query a PostgreSQL table in real time to get features such as a user’s credit score.

(This feature is currently in private preview. If you’re interested in participating, file a support ticket.)

Read more about API Resources →

Speed up real-time execution with faster, simpler expressions

Tecton is also introducing a new capability that makes it even more efficient to perform the calculations needed for Realtime Feature Views, which can speed up transformations during online retrieval queries. Simple operations such as basic arithmetic, default values, or date-diffs can now be defined directly in the features definition using the new Feature type Calculation.

Calculations define feature logic through simple SQL-like expressions, bypassing user-defined Python functions in traditional Realtime Feature Views. Calculation features can be an efficient alternative for Feature Views that don’t require the expressiveness of a python or pandas mode transformation.

(This feature is currently in private preview. If you’re interested in participating, file a support ticket.)

Read more about Calculations →

Enhance performance with core platform improvements

Tecton continues innovating our core platform to make feature production more cost-optimized and efficient. Here are the latest:

• Replication Constraint in Dynamo Config: Use the new replica_region parameter in the DynamoConfig for a Feature View’s online store to specify which regions you want to constrain replication to. This allows you to avoid costly replication charges to your satellite regions that you don’t need to serve traffic to.
• Migrate off of Python 3.7 in EMR: You can now run EMR compute on Python 3.9 regardless of which EMR version you use. This allows users running in EMR version 6.X to get off of Python 3.7, which reached end of life in June 2023 and is no longer receiving security upgrades.
• Improved Unit Testing Support: Using the MockContext class, you can now set custom mock secrets and resources for your queries, enabling faster development and iteration in unit test and notebook environments. View the Unit Testing docs for more details.

To upgrade to 1.1, follow the upgrade guide here. And if you’re interested in trying out the new features in private preview, go ahead and file a support ticket. We’re excited to see what our customers will build next with these new capabilities!