This anthropic article made me think of several InterSystems presentations and articles on the topic of data quality for AI applications. InterSystems is right that data quality is crucial for AI, but I imagined there would be room for small errors, but this study suggests otherwise. That small errors can lead to big hallucinations. What do you think of this? And how can InterSystems technology help?

https://www.anthropic.com/research/small-samples-poison

The ObjectScript language has incredible JSON support through classes like %DynamicObject and %JSON.Adaptor. This support is due to the JSON format's immense popularity over the previous dominance of XML. JSON brought less verbosity to data representation and increased readability for humans who needed to interpret JSON content. To further reduce verbosity and increase readability, the YAML format was created. The very easy-to-read YAML format quickly became the most popular format for representing configurations and parameterizations, due to its readability and minimal verbosity.

The built-in task manager is limited. How can I implement a robust, distributed job scheduler in IRIS with support for dependencies, CRON syntax, and failover recovery?

We require automatic injection of security predicates at runtime, depending on the user or API token. Is there a supported or hackable mechanism to manipulate SQL parsing/compilation before execution?

We use Apache Flink for complex event processing. Is there a way to integrate IRIS (as a source/sink) with Flink’s streaming API, possibly using the IRIS Native API or JDBC?

We’re encountering occasional deadlocks when accessing persistent objects. How can I trace lock acquisition and identify cyclic dependencies in real time?

I have large joins involving millions of rows. How can I profile and tune the SQL engine’s parallel execution? Are there EXPLAIN plan features to inspect threading and task distribution?

When writing dynamic SQL queries using embedded SQL, how can I force or ensure that filter conditions are pushed down to the data access layer rather than evaluated in memory?

We want to iterate over large global structures in real-time without blocking or locking readers. How can we safely use $Order() and implement a lock-free analytics approach?

What are the best practices for creating a multi-tenant app in IRIS? How can I isolate data per tenant using namespaces, control resource usage, and delegate access via roles securely?

For space optimization, we want to apply a domain-specific compression algorithm to binary stream data before writing to %Stream.GlobalBinary. Is it possible to override or extend stream classes to include compression/decompression?

Deploying new IRIS instances can be a time-consuming task, especially when setting up multiple environments with mirrored configurations.

I’ve encountered this issue many times and want to share my experience and recommendations for using Ansible to streamline the IRIS installation process. My approach also includes handling additional tasks typically performed before and after installing IRIS.

This code provide the configured production items with enabled or disabled status.

Include (Ensemble, EnsUI, EnsUtil)

Class Test.ProductionConfig
{

Hello everybody,

I would like to export project contents using Visual Studio Code, as well as I do with InterSystems Studio.

However, while attempting the export through the InterSystems extension I get the following error:

- There are no folders in the current workspace that code can be exported to.

These are the step to reproduce my error: