The goal of this residency is to update REDP-5661-00, Optimized Inferencing and Integration with AI on IBM Z: Introduction, Methodology, and Use Cases. The update will include: Updates to the credit risk scoring on IBM Z use case to discuss the full product of WMLz V2.4 in further detail, highlighting the batching capability it provides and the value add it brings when compared to WMLz OSCE when AIU is present, highlight the CICS Cobol Scoring service from COBOL language programs, and describe and demonstrate the steps to modify the implementation to leverage AIU. It will also imclude the addition of a new chapter that discusses and demonstrates the implementation and integration of an end-to-end solution leveraging Deep Learning models to enforce a Payments Fraud Detection use case, based on an internal demo, and discuss the benefits derived from the solution, focusing on the value derrived from leveraging AIU and IBM Z.
Starts 12 Sep 2022, ends 17 Oct 2022 (5 weeks) and requires 5 residents
Residency Leader: Makenzie Manna
Benefits to Resident
IBM Redbooks want technical practitioners to be recognized for their newly gained knowledge, experience, and accomplishments. Therefore, upon successful delivery of the publication produced by this IBM Redbooks residency, you will become eligible to earn an IBM Redbooks Digital Badge issued through Credly.
Update credit risk scoring use case:
- Document the benefits received from leveraging the full WMLz product versus WMLz OSCE when the z16 on-chip AI accelerator is present, highlighting the batching capability and CICS COBOL Scoring service from COBOL language programs.
- Describe the steps to modify the implementation to leverage AIU.
Add a new chapter discussing a Payments use case on IBM Z that targets bank account to bank account transfer payments fraud (based on an internal banking showcase demo):
- Discuss the business value of the solution with a focus on the unique benefits provided by Z.
- Provide examples of industry scenarios that can leverage the solution.
- Describe the use case.
- Illustrate and describe reference architecture for the solution and how the pieces of the solution fit together and how the software components take benefit of the on-chip AI acceleration.
- Describe the data science process of understanding and preparing the data for the model.
- Describe the process of (Long Short-Term) model development and training.
- Describe the process of (Long Short-Term) Model deployment and evaluation on IBM Z.
*Residents will be responsible for verifying the system environment setup and applying any updates, reconfigurations, or configuring new software to enable the implementation.
A basic requirement for all residents is the ability to read and clearly express concepts and procedures in common English.
|Positive, self-motivated, professional||5|
|Teamwork and leadership||5|
|Python skills for data science||4|
|Data Science skills to demonstrate how to prepare and use data for AI (select model type, build and train model)||4|
|Deploy models for inferencing on IBM Z using WMLz and leveraging AIU||4|
|Benefits of full WMLz V2.4 product versus WMLz OSCE when AIU is present||4|
|Red Hat OpenShift Container Platform on IBM Z||3|
|Db2 for z/OS skills||4|
|Solution architecture and integration knowledge on IBM Z||4|
|Fraud detection solutions on IBM Z||4|
|CICS COBOL Scoring service from COBOL language programs in WMLz||3|
|Technical writing experience||4|
|PC-based Graphics Tools (PowerPoint, Visio, Adobe Illustrator)||3|
Social Media Skills not specified