What the Dev?

In this episode, David Rubinstein interviews, Stephen Watt, VP of the Office of the CTO at Red Hat, about open source AI.

They discuss: 

• Differences between open source software and open source AI
• The difficulty in applying open source principals to AI
• Efforts made to define open source AI

In this episode, Jenna Barron interviews Martin Reynolds, field CTO of Harness, about the misalignment between developers and FinOps. The company recently did a report that backed up the claims that developers and FinOps aren't properly aligned, and they talk about some of the findings of that report. 

Some key points of discussion include:

• The fact that over half of developers say their cloud purchasing commitments are based on guesswork
• The lack of data on idle cloud resources, unused resource, and over or underprovisioned workloads
• Processes and tools that can help with alignment
• How the adoption of AI has contributed to the problem

Read Harness' report here: https://www.harness.io/finops-in-focus 

Read the transcript of this episode here: https://sdtimes.com/softwaredev/qa-the-growing-divide-between-developers-and-finops/ 

In this episode, Jenna Barron interviews Lukasz Karolewski, senior staff software engineer at LinkedIn, and Ajay Prakash, staff software engineer at LinkedIn, to talk about how they built a prompt engineering playground at LinkedIn to improve cross-team collaboration with non-technical teams.

They discuss:

• Why they built the playground
• The choice to use Jupyter notebooks
• The work they did to make sure it was accessible to non-technical users
• How it's facilitated faster innovation at the company

They detailed their work in this blog post here: https://www.linkedin.com/blog/engineering/product-design/building-collaborative-prompt-engineering-playgrounds-using-jupyter-notebook 

In this episode David Rubinstein interviews Eran Yahav, co-founder and CTO of Tabnine, about why its important to provide AI assistants with clean code.

They discuss: 

• The importance of defining organizational rules and best practices to guide the AI
• How to use AI-assisted refactoring to improve legacy codebases
• If organizations should apply KonMari principles to their codebases

Join SD Times news editor Jenna Barron for a panel discussion on DeepSeek and what developers need to know about this model that has been in the spotlight for the past few weeks. 

The conversation covers why DeepSeek has gained so much attention, what makes it different from other models, proper data security and hygiene practices for using DeepSeek, and predictions for the future of AI innovation. 

The speakers include:

• Melissa Ruzzi, director of AI at AppOmni
• Bratin Saha, chief product and technology officer at DigitalOcean
• Kate O’Neill, author, speaker, and executive consultant on technology’s impact on the human experience

Related resources from our speakers:

• Kate’s new book, What Matters Next: A Guide to Making Human-Friendly Tech Decisions - https://www.koinsights.com/books/what-matters-next-book/
• AppOmni’s webinar on AI security and balancing the risks and benefits of AI - https://appomni.com/webinars/ai-security-balancing-business-and-saas-risks/ 
• AskOmni, a chatbot for SaaS security - https://appomni.com/askomni/ 
• How to Run DeepSeek R1 Large Language Models on DigitalOcean GPU Droplets - https://www.digitalocean.com/community/tutorials/deepseek-r1-gpu-droplets 
• DigitalOcean Deploy 2025 Keynote - https://www.youtube.com/watch?v=phxuIai5cD8 

The video for this conversation was also posted to our YouTube channel: https://www.youtube.com/watch?v=Zdtkj2YN4Nw 

In this episode, David Rubinstein interviews Damien Garros, co-founder and CEO of OpsMill, about middle code, a concept that sits somewhere between programming and low-code/no-code.

They discuss:

• What middle code is
• The influence generative AI has had on this technology
• How middle code can be used to accomplish things not possible with low-code or no-code

In this episode, we interview Dylan Thomas, senior director of product engineering at OpenText Cybersecurity, about the evolution from shift left to shift everywhere.

At the end of 2024, he predicted: "In 2025, DevSecOps will continue evolving beyond the ‘shift-left’ paradigm, embracing a more mature ‘shift everywhere’ approach. This shift calls on organizations to apply the right tools at the right stages of the DevSecOps cycle, improving efficiency and effectiveness in security practices. Lightweight analysis in IDEs will help developers catch issues early, while automation integrated into pull requests and CI/CD pipelines will ensure a cohesive ‘integrate once’ approach for core functions such as SAST, SCA, and increasingly DAST, particularly for API security testing."

We interviewed him about his predictions, and talked about: 

• What shift everywhere is
• Why people are wanting to transition to this new approach
• How to get started with shift everywhere

In this episode, we spoke with Trisha Gee, lead developer advocate at Gradle, about flaky tests and the challenges of dealing with them.

Key talking points:

• What is a flaky test?
• How can they be identified?
• How can they be prevented in the future? 

Other useful resources on flaky tests: 

• Develocity’s Flaky Test Detection
• Trisha’s recent Flaky Test webinar 
• Trisha’s Blog - How Top Banks Handle Flaky Tests
• Desirable Unit Tests By Kent Beck
• Martin Fowler's Definition of a Contract Test 
• Your Tests are Failing You! (video) 
• 5 Reasons Your Automated Tests Fail (video) 

In this episode, we spoke with Derek Holt, CEO of Digital.ai, about the emerging practice of software engineering intelligence (SEI). At the end of last year, he predicted that "SEI will have a breakout year in 2025 as more and more businesses realize they need to measure the end-to-end business process of software development and delivery in order to drive continuous improvement, truly deliver improved developer experiences and ultimately realize the potential gains for an AI-powered Software Development and Delivery capabilities."

We invited him on the podcast to dive into that prediction further, and we discussed:

• What SEI is
• Why it's gaining traction this year
• How it differs from value stream management, and why it will be more successful

Read all 2025 predictions here: https://sdtimes.com/softwaredev/predictions-for-software-development-in-2025/

In this episode, we interview Scott McCarty, senior principal product manager of Red Hat Enterprise Linux, about what we can look forward to in 2025 when it comes to containers. 

Key talking points include:

• Cross-platform use cases for containers
• Bending containers to do more for us
• An overview of bootc and how it fits into the future of containers