Apply to Test the Infineon PSOC™ Edge E84 AI Evaluation Kit

View table of contents ...  

This RoadTest gives you the opportunity to try the PSOCTm Edge E84 AI Evaluation Kitkit, use DEEPCRAFTTm Studio and DEEPCRAFTTm Ready Models, and evaluate how well it supports AI and ML workloads at the edge.

The Infineon PSOCTm Edge E84 AI Evaluation Kit (KITPSE84AITOBO1) is built for testing machine learning and Edge AI applications. It’s designed to help engineers and developers explore Infineon’s DEEPCRAFTTm AI suite, collect data from onboard sensors, and experiment with AI models directly on the device.

What You Will Be Testing

  • The out-of-box experience — from setup to running your first example.

  • Using DEEPCRAFTTm Studio to collect data from onboard sensors.

  • Running and evaluating DEEPCRAFTTm Ready Models (for example, baby cry detection or motion type detection).

  • Developing code with ModusToolboxTm software.

  • Debugging and programming using the on-board KitProg3 interface.

You’ll share your experience through text, images, or video, showing other community members how easy it is to get started and what the kit can do.

What’s Included

RoadTesters will receive or have access to:

  • Infineon PSOCTm Edge E84 AI Evaluation Kit

  • ModusToolboxTm software (v3.6 or later)

  • DEEPCRAFTTm AI Suite, including:

    • DEEPCRAFTTm Studio

    • DEEPCRAFTTm Ready Models

Key Features

  • Cortex®-M55 CPU with HeliumTm DSP and EthosTm-U55 NPU for AI acceleration

  • NNLite NPU for ultra-low-power ML processing

  • Multiple onboard sensors:

    • 60GHz XENSIVTm radar

    • Barometric pressure sensor

    • Magnetometer

    • Analog and digital microphones

    • Image sensor

  • Built-in wireless (Wi-Fi + Bluetooth®)

  • On-board KitProg3 debugger

  • USB host/device interfaces and MIPI-DSI display connector

RoadTester Tasks and Instructions

  1. Unbox and explore the kit, following the Getting Started Guide.

  2. Test the out-of-box setup and initial software configuration.

  3. Collect data using DEEPCRAFTTm Studio and onboard sensors.

  4. Flash and test a DEEPCRAFTTm Ready Model (such as baby cry or gesture classification).

  5. Develop and debug code using ModusToolboxTm and explore example projects.

  6. Share your findings in a review, including what worked well and any challenges.

The preferred way to test it would be to use it with something you are currently working on. Otherwise, working with the documentation, open the box and test the out of the box experience. Show the reader of this review what you discovered utilizing text, images, videos, and/or any other media, so the reader understands what is involved in using the product.

Steps to Test DEEPCRAFT Studio and DEEPCRAFT Ready models

  • Flash and use the Data Collection CE to collect data to a model in DEEPCRAFTTm Studio (we suggest MovementTypeDetection to start with)
  • Collect Image data using the Data Collection Object Detection project in DEEPCRAFTTm Studio.
  • Download the evaluation version of one of the DEEPCRAFTTm Ready Models (we suggest the DEEPCRAFTTm Ready Model for Baby Cry Detection)
  • Flash the evaluation version of the DEEPCRAFTTm Ready Model to your device and run live tests in your desired environment

Useful Links

Important Dates

Begin enrollment

October 16 2025

End enrollment

November 16 2025

Select RoadTesters

Week of 17th November

Ship unit

Week of 24th November

Begin RoadTesting

On Receipt of Kit

element14 followup

6 weeks after Selection

Post reviews by

2 to 3 month after Kit Receipt

Terms and Conditions

About the Sponsor

Infineon Technologies is a leading manufacturer of semiconductor and system solutions for automotive, industrial electronics, chip card and security as well as applications in communications with a strategic focus on the three main areas: Energy Efficiency, Communications and Security Infineon’s products stand out for their reliability, their quality excellence and their innovative and leading-edge technology in analog and mixed signal, RF and power as well as embedded control. A strong technology portfolio with about 22,900 patents and applications is characteristic for the company. For more information, click here.

RoadTest Reviews
Comment List
Anonymous