Kevin Kirkus on Automated Testing Embedded Code - Episode 186

Kevin Kirkus is a Principal Engineer at Intel. He has been there since 1999, designing, building, and testing Intel processors, both the chip and the code that runs the chips. Kevin is a Post-Silicon Validation Architect Lead for multiple Intel Xeon generations and more recently transitioned to Global Post-Silicon Automation Lead for server, client, and device products at Intel Corporation. He has served in various past areas of validation, such as PCIe, socket-interconnect, Platform Configurations, Boot Flows, Fuse, before transitioning to more technical leadership roles in validation architecture and strategy. As a Technical Leader in silicon validation automation, his roles and responsibilities have extended to global standardization and alignment of tools, flows, and methodologies across Intel global product segments (server, client, device), including mentoring environment domain, leads, aligning global sites and strengthening local product sites to excel in automation investments. His technical and leadership experience for validating advanced complex technology has been integral to the success of Intel product teams and their competitive products. He is passionate about developing people and future leaders, building on their strengths, influence, and impact.

Topics of Discussion:

[3:33] What are some high points that lead Kevin to where he is today?

[8:06] Kevin talks about focusing on functional validation execution, and what he likes as a Post-Silicon Validation Architect Lead.

[9:03] Kevin walks us through the process from where the code is embedded, the silicon is made, and there's a system of trial and error to test if you are on the right track.

[15:33] What are the penalties for the turnaround time?

[15:48] What is the mix of code types that are a part of this?

[17:18] Kevin talks about the concept of fuses and SKUs or stock-keeping units.

[21:12] Kevin talks about the format of the test themselves, the language in which he writes the test cases, and the partnership between the host and target system.

[31:42] Where are test harness codes and firmware loaders stored?

[31:47] Firmware has to be handled a little bit differently; it has to be cross-platform deployment packaging.

[35:20] Most of Intel Xeon's solutions are proprietary, and their bug tracking is proprietary.

[39:05] How does Kevin coach people to have the technical skills they need in this space?

Mentioned in this Episode:

Architect Tips — New video podcast!

Azure DevOps

Clear Measure (Sponsor)

.NET DevOps for Azure: A Developer's Guide to DevOps Architecture the Right Way, by Jeffrey Palermo — Available on Amazon!

Jeffrey Palermo's YouTube

Jeffrey Palermo's Twitter — Follow to stay informed about future events!

Want to Learn More?

Visit AzureDevOps.Show for show notes and additional episodes.

Quotes:

  • "My job starts when they have gotten to the point where it boots up and it's reaching the point of fetching the operating system code to where we can actually see it in the operating system." — Kevin [20:02]
  • "If we can deliver a healthy environment to our functional validation team, it gives them a chance to focus their energies on writing tests that actually find the most bugs." — Kevin [39:54]

Kevin Kirkus: LinkedIn

Intel

Intel Xeon


Download this episode (MP3, 47:09, 45.3 MB)

Watch this episode on YouTube

This episode on the AI DevOps Podcast site

Sam Nasr on Azure AI & ML - Episode 185

This week, Sam Nasr returns to the show. Sam is an IT Consultant specializing in .Net, SQL Server, and Azure. He is a Sr. Software Engineer focused on the Microsoft stack of technologies including .Net, SQL Server, Azure. As part of NIS Technologies, he provides consulting services, training, and custom app development to bring more value to business applications. Sam is also a leader at the Cleveland C#/VB .Net user group.

Topics of Discussion:

[2:37] Sam talks about getting inspiration from a video featuring Saquib Shaikh, a blind software engineer at Microsoft that developed services that helped him get by day-to-day, which was made public. It was a moving video and it got him into cognitive services, and he started diving more into AI and ML from that point.

[4:10] How are AI and machine learning similar? How are they different?

[4:15] Do you have to use the Azure services?

[6:27] What are some of Sam's favorite AI or ML products?

[8:03] With Azure Cognitive Services, it's a variety of different services. There are some for language, some for speech, some for vision, and decision-making.

[10:19] How do we go from a microphone to a text string? Does that happen on the end device, or do we send the file to Azure?

[15:10] What is the testing method for this type of application?

[18:09] How does this relate to bot framework?

[24:28] What applications does it tend to work best on?

[26:50] What does it look like to deploy a new release candidate from environment to environment and then to production? What is the promotion process of an application like this?

[30:39] You need to have a good representation of your data. It's got to be clean and then trained on an appropriate number of records.

Mentioned in this Episode:

Architect Tips — New video podcast!

Azure DevOps

Clear Measure (Sponsor)

.NET DevOps for Azure: A Developer's Guide to DevOps Architecture the Right Way, by Jeffrey Palermo — Available on Amazon!

Jeffrey Palermo's YouTube

Jeffrey Palermo's Twitter — Follow to stay informed about future events!

Want to Learn More?

Visit AzureDevOps.Show for show notes and additional episodes.

Sam Naser:

GitHub projects

YouTube

Presentations

samnasr@live.com

Overview of Cognitive Services:

"Adding Machine Learning to .Net Applications"

youtube.com/watch?v=dMHlbZvISUI

Build 2016 AI Video


Download this episode (MP3, 35:19, 33.9 MB)

Watch this episode on YouTube

This episode on the AI DevOps Podcast site

Chris Tacke on .NET 6 IoT on Linux - Episode 184

Chris Tacke is an industry leader in managed application development for industrial process control, medical, telematics, and just about any other embedded industry. Chris specializes in Windows CE and the .NET Compact Framework, and mobile and embedded device application development.

Topics of Discussion:

[2:22] Fun fact! Chris got a degree in Geology long ago, and his first job was doing seismic work and hand-analyzing a lot of data. Chris talks about his first foray into programming, a job in Excel, writing VBA macros to analyze the data.

[4:08] Chris walks us through a job at Microsoft by way of a horse race in Virginia during the .com boom. After the .com crash, he worked at a company that made single board computers and moved to be an engineering manager for the Windows CE department. He was independent for 20+ years before joining Hypergiant, and then Wilderness Labs came onto his radar.

[6:42] Chris is the owner of the Meadow Core pieces and does a fair bit of the driver development.

[8:49] How does the Jetson Nano compare to Raspberry Pi?

[13:47] Chris talks about the first goal to run on modern.net.

[16:41] What is the vision now to say that we need to do this on Linux?

[21:33] Is this the same vision as Maui, to run anywhere embedded?

[23:38] How does it impact the process of building, test suites, and different test environments?

[24:18] Chris talks about Meadow Test Suite, which has the goal to deploy binaries down to a device and find tests that look just like an xUnit test that can then talk to hardware.

[30:56] This totally redefines the term of the "full-stack" developer.

[31:12] What can people utilize today, and what is the DevOps chain Chris and his team are using?

Mentioned in this Episode:

Architect Tips — New video podcast!

Azure DevOps

Clear Measure (Sponsor)

.NET DevOps for Azure: A Developer's Guide to DevOps Architecture the Right Way, by Jeffrey Palermo — Available on Amazon!

Jeffrey Palermo's YouTube

Jeffrey Palermo's Twitter — Follow to stay informed about future events!

Wilderness Labs Dev Camp

ClassMeadowsOS

Wilderness Labs — Github

Jetson Xavier NX Series

Want to Learn More?

Visit AzureDevOps.Show for show notes and additional episodes.


Download this episode (MP3, 39:24, 37.8 MB)

Watch this episode on YouTube

This episode on the AI DevOps Podcast site

Henry Quillin on Prepping for a Career as a Software Engineer - Episode 183

Henry Quillin is a high school senior interested in software development, entrepreneurship, and blockchain/crypto. He has completed several internships and other contracts and recently earned the rank of Eagle Scout in the Boy Scouts of America. He always has several software projects going including ProjectNEWM, an attempt to decentralize the music industry, and when not buried in VSCode or books, he enjoys weightlifting, cooking, and listening to podcasts. You can check out his website at henry quillin.me.

Topics of Discussion:

[2:39] Henry talks about becoming interested in software development and programming as early as 4th grade. While his friends were playing video games, he was creating them.

[5:00] Henry discusses his first shadowing experience that ended up turning into a paid internship at Jackson & Ryan Architects.

[10:10] What was it like for Henry to see the development process from a bird's eye view during his first paid internship in tech at a startup called Original Nations?

[10:47] How did Henry get into Clear Measure?

[12:18] Henry talks about the huge community in programming that wants to help and how they have connected him with different opportunities.

[12:55] What is projectNEWM? How is it like a decentralized Spotify?

[16:51] Henry has some sound advice for fellow young adults his age looking to get into the programming industry, and what he feels the benefits of internships are.

[27:06] What is the value that Henry sees in college, and which schools are his top choices?

[29:05] Henry feels that if you want to become a developer, just really try and get some experience.

Mentioned in this Episode:

Architect Tips — New video podcast!

Azure DevOps

Clear Measure (Sponsor)

.NET DevOps for Azure: A Developer's Guide to DevOps Architecture the Right Way, by Jeffrey Palermo — Available on Amazon!

Jeffrey Palermo's YouTube

Jeffrey Palermo's Twitter — Follow to stay informed about future events!

ProjectNEWM

Henry Quillin

Want to Learn More?

Visit AzureDevOps.Show for show notes and additional episodes.


Download this episode (MP3, 38:59, 37.4 MB)

Watch this episode on YouTube

This episode on the AI DevOps Podcast site