Software Sundays

Software Sundays is a weekly podcast where technology, culture, and real-world impact intersect. Hosted by Kevin Dowdy, the show explores the latest trends in software engineering, AI, and digital innovation—while breaking down what they actually mean for engineers, builders, and communities. From industry shifts to practical insights, each episode is designed to help you think critically, build intentionally, and lead with purpose. Whether you're a developer, founder, or someone looking to transition into tech, this is your space to stay informed and grow.

Software Sundays is a news podcast hosted by Kevin Dowdy, with 22 episodes on record and a Required Pod Score of 80.

Kevin Dowdy hosts Software Sundays, a news show with 22 episodes published.

Our AI reads these to draft pitches. Use them as grounding for a pitch that cites a real guest and a specific topic.

• 

  Episode #44

  AI Could Replace Your Job, But That’s Not the Real Risk | Software Sundays | Software Sundays #44

  Sep 27, 202642 min

  This week on Software Sundays, KD breaks down the growing fear around artificial intelligence and why engineers and builders need to think beyond the headlines. We start with warnings from AI researchers about the potential risks of increasingly capable models, recent incidents involving frontier AI systems, and the lack of transparency surrounding how these models are developed. But instead of focusing only on worst-case scenarios, KD looks at the bigger question: who is building, deploying, and ultimately responsible for these systems? AI development isn’t stopping. Billions are being invested into the models, infrastructure, and data centers powering the next generation of AI. The opportunity for builders is learning how to understand these systems, deploy them responsibly, and determine where AI should operate autonomously versus where human judgment still matters. Then we get into this week’s engineering Q&A. KD breaks down what great performance actually looks like as a software engineer, why helping your team improve can be more valuable than individual output, and whether AI will eventually replace your job. We also cover where someone should start if they want to become a software engineer today, from choosing a programming language to moving beyond tutorials and building real projects. KD explains what accountability means when you own a project and shares a practical framework for prioritizing planned work, urgent requests, and long-term improvements. We close with a reminder that discipline matters, but your life shouldn’t become so structured that you stop experiencing anything new. Sometimes stepping outside the routine gives you the energy you need to keep going. Timestamps: 00:00 - Welcome to Software Sundays 00:30 - AI researchers are warning about catastrophic risk 09:44 - What does great software engineering performance look like? 14:20 - Will AI replace your job? 19:21 - How to become a software engineer today 24:08 - What does project accountability really mean? 28:26 - How engineers should prioritize their work 36:08 - Why disciplined people still need spontaneity 39:14 - Outro Join BLI University: https://discord.gg/jpJHGq6kgS Connect with Build Learn Impact: Instagram → https://www.instagram.com/build.learn.impact Substack → https://buildlearnimpact.substack.com Connect with Kevin Dowdy: Instagram → https://www.instagram.com/mrkevindowdy LinkedIn → https://www.linkedin.com/in/kevinldowdy Substack → https://substack.com/@kevinldowdy GitHub → https://github.com/kevindowdy DISCLAIMER: This is not professional advice. The views are my own and the people quoted. Consult your own advisers for legal, business or tax decisions based upon information from this episode. Build Learn Impact is on a mission to help you create wealth and opportunity through technology. Subscribe if you’re ready to build the future. #SoftwareSundays #AI #ArtificialIntelligence #SoftwareEngineering #TechCareers #AIJobs #Engineering #Programming #CareerDevelopment #TechEducation #Leadership #SoftwareDeveloper #BuildLearnImpact

  ListenShow Notes
• 

  Episode #43

  DoorDash’s AI Agent Pipeline, Git Worktrees & When Engineers Should Build Their Own Tools

  Sep 20, 202638 min

  This week on Software Sundays, KD breaks down how DoorDash is using a multi-agent LLM pipeline to clean up more than 60,000 stale feature flags across hundreds of repositories — and why AI agents may become increasingly valuable for the engineering work teams know needs to get done but rarely have enough time to prioritize. Then we get into how engineering teams can actually burn down large backlogs faster than new work is created. KD covers reducing false positives, improving triage, prioritizing high-impact fixes, increasing development capacity, and why separate development environments and faster QA processes are critical for moving quickly without creating unnecessary risk. In this episode’s Q&A, KD explains why Git worktrees have become especially useful in the age of AI coding agents, how they allow multiple developers or agents to work in parallel without stepping on each other’s changes, and how they can fit into an existing branch and pull request workflow. We also break down what Datadog is used for, how logs and metrics support observability, how Datadog agents collect telemetry, and how teams can build dashboards and alerts around the systems they operate. KD also covers how to manage feature flags correctly — including choosing safe defaults, connecting flags to external systems like LaunchDarkly, monitoring rollouts, establishing rollback criteria, and making sure observability is in place before experimenting with production behavior. For this week’s engineering tip, KD challenges the idea that engineers should never reinvent the wheel. After needing better visibility into automated data pipelines, he built a custom log forwarder for Datadog rather than going through the process of deploying third-party software in a restricted environment. The lesson: build vs. buy decisions should depend on your constraints, the value of the solution, and what you learn from building it yourself. We close with a reminder that growth often requires putting yourself in rooms with people who are better than you. Whether you're developing as an engineer, architect, security professional, entrepreneur, or leader, your environment can either challenge you to improve or keep you comfortable at your current level. Chapters: 00:00 DoorDash's Multi-Agent Feature Flag Cleanup LLM 01:37 Using AI to Manage Stale Feature Flags 05:06 Handling Backlogs and Accelerating Development 07:26 Strategies for Reducing False Positives in Backlog Management 12:15 Introduction to Git Work Trees for Parallel Development 17:37 Monitoring with Datadog: Logs, Metrics, and Alerts 21:44 Best Practices for Feature Flag Management 25:15 Using Monitoring to Decide When to Roll Back Features 26:52 Engineering Tips: Building vs Buying Solutions 27:10 Engineering tip of the week 33:09 Don't be the smartest in the room ------------------------- Join BLI University: https://discord.gg/PFH8Qszk2A Connect with Build Learn Impact: Instagram → https://www.instagram.com/build.learn.impact Substack → https://buildlearnimpact.substack.com Connect with Kevin Dowdy: Instagram → https://www.instagram.com/mrkevindowdy LinkedIn → https://www.linkedin.com/in/kevinldowdy Substack → https://substack.com/@kevinldowdy GitHub → https://github.com/kevindowdy DISCLAIMER: This is not professional advice. The views are my own and the people quoted. Consult your own legal, business or tax advisors before making decisions based upon information found in this episode. Build Learn Impact is on a mission to help our community create wealth and opportunity through technology. Subscribe if you’re ready to build the future.

  ListenShow Notes
• 

  Episode #42

  OpenAI’s AI Agent Escaped Its Sandbox, Found Zero-Days & Started a Swarm | Software Sundays

  Sep 13, 202644 min

  This week on Software Sundays, KD breaks down a cybersecurity experiment that shows just how unpredictable AI agents can become when they are given a goal and enough room to operate. We start with an OpenAI cybersecurity evaluation where a sandboxed AI agent found a path outside of its restricted environment, chained together vulnerabilities across multiple systems, discovered zero-days, and even found ways to collaborate with other agents. KD unpacks what happened, why traditional sandboxing was not enough, and what this could mean as AI agents become more capable. From there, we get into how sandbox environments should actually be configured, including production isolation, network segmentation, infrastructure as code, monitoring, cost controls, and treating experimental environments as disposable infrastructure. In this episode’s Q&A, KD explains what a zero-day vulnerability is, why defense in depth matters, what cybersecurity engineers actually do, and the difference between red teams and blue teams. We also get into the growing distinction between vibe coding and agentic engineering. AI can write code quickly, but engineering still requires architecture, guardrails, design decisions, monitoring, and someone capable of recognizing when the system is going in the wrong direction. As AI-powered attacks begin operating at machine speed, defenders will increasingly need automation that can detect, isolate, respond to, and recover from threats just as quickly. We close with a reminder to seek quiet moments in an increasingly loud world, make space for your own voice, and be intentional about which voices you allow to influence your direction. Chapters: 00:00 Introduction to the episode and main themes 00:27 Recent AI cybersecurity incident between OpenAI and Hugging Face 10:17 Configuring secure sandbox environments for AI development 18:11 Understanding zero-day vulnerabilities and their risks 24:07 Role of cybersecurity engineers and their responsibilities 28:12 Vibe coding versus agentic engineering in AI development 32:30 Red team and blue team roles in cybersecurity defense 38:46 The importance of internal voice and mindfulness in a noisy world Join BLI University: https://discord.gg/PFH8Qszk2A Connect with Build Learn Impact: Instagram → https://www.instagram.com/build.learn.impact Substack → https://buildlearnimpact.substack.com Connect with Kevin Dowdy: Instagram → https://www.instagram.com/mrkevindowdy LinkedIn → https://www.linkedin.com/in/kevinldowdy Substack → https://substack.com/@kevinldowdy GitHub → https://github.com/kevindowdy DISCLAIMER: This is not professional advice. The views are my own and the people quoted. Consult your own advisers for legal, business or tax decisions based upon information from this episode. Build Learn Impact is on a mission to help our community create wealth and opportunity through technology. Subscribe if you’re ready to build the future.

  ListenShow Notes
• 

  Episode #41

  Stablecoins Are Becoming Infrastructure & Engineers Need to Be Ready | Software Sundays #41

  Sep 6, 202630 min

  This week on Software Sundays, KD breaks down how stablecoins are becoming part of the financial infrastructure businesses and financial institutions rely on — and why that shift is creating major opportunities for software engineers and builders. We get into what it actually takes to connect stablecoin products with existing financial rails, including APIs, authentication, treasury infrastructure, cybersecurity, compliance, fraud detection, and the engineering required to make faster and more global payments reliable. KD also explains why the next generation of financial technology will still need to integrate with legacy systems, and why understanding both software architecture and the regulated environments these systems operate in can make engineers more valuable as finance continues to evolve. In this episode’s Q&A, we cover how to manage a release cycle for a production application, why changes should be classified by risk, how canary deployments and rollback strategies reduce production failures, and why mission-critical systems require careful release planning. We also break down how to enable payments inside an application without rebuilding financial infrastructure from scratch, why services like Stripe can reduce complexity and compliance risk, which careers benefit from a software engineering background, and what a single point of failure means when designing reliable systems. We close with a reminder about the value of direct communication: sometimes the shortest path to solving a problem is simply saying clearly what needs to be said. Chapters: 00:28 STABLECOIINS ARE BECOMING PART OF THE FINANCIAL INFRASTRUCTURE 08:08 MANAGING RELEASE CYCLES FOR ENTERPRISE APPLICATIONS 14:58 ENABLING PAYMENTS WITHIN YOUR APPLICATION USING THIRD-PARTY SERVICES 17:41 THE VALUE OF A SOFTWARE ENGINEERING BACKGROUND IN FINANCE AND OTHER INDUSTRIES 23:06 UNDERSTANDING AND MITIGATING SINGLE POINTS OF FAILURE IN SYSTEMS 25:46 EFFECTIVE COMMUNICATION AND DIRECTNESS IN TEAM MANAGEMENT Join BLI University: https://discord.gg/wvsWhZ3A Connect with Build Learn Impact: Instagram → https://www.instagram.com/build.learn.impact Substack → https://buildlearnimpact.substack.com Connect with Kevin Dowdy: Instagram → https://www.instagram.com/mrkevindowdy LinkedIn → https://www.linkedin.com/in/kevinldowdy Substack → https://substack.com/@kevinldowdy GitHub → https://github.com/kevindowdy DISCLAIMER: This is not professional advice. The views are my own and the people quoted. Consult your own advisers for legal, business or tax decisions based upon information from this episode. Build Learn Impact is on a mission to help our community create wealth and opportunity through technology. Subscribe if you’re ready to build the future.

  ListenShow Notes
• 

  Episode #40

  The EU AI Act Is Here: High-Risk AI, Security & Human Oversight | Software Sundays #40

  Aug 30, 202642 min

  This week on Software Sundays, KD breaks down what happens when AI regulation moves from theory into enforcement. Article 50 of the EU AI Act is now in effect, bringing new transparency requirements for companies deploying AI systems. We get into what businesses need to disclose, how the European approach differs from the largely voluntary AI frameworks in the United States, and why companies operating across multiple markets may eventually build around the strictest standard. Then we get into the practical side of AI risk. KD explains how to test the security of an AI system, why prompt injection remains a major concern, and why AI cannot be trusted to police itself. We cover access controls, external APIs, sensitive data, software dependencies, penetration testing, and the additional safeguards teams need when putting autonomous systems into production. In this episode’s Q&A, we also break down the difference between a data controller and data processor, how AWS S3 lifecycle policies can support regulatory compliance while controlling storage costs, and how to determine whether an AI system should be considered high risk. We also unpack human-in-the-loop, human-on-the-loop, and human-out-of-the-loop architectures — and why the amount of human oversight you need should increase alongside the potential impact of an AI decision. We close with a mindset reminder about risk: being prepared does not mean being paralyzed. Understand what could go wrong, build safeguards, and keep moving toward the opportunities worth exploring. Chapters: 00:00 Introduction to AI regulation and the EU AI Act 01:56 Overview of the EU AI Act's scope and requirements 03:42 Comparison between EU regulations and US approach 07:59 Implications for companies operating inside and outside the EU 08:54 Security vulnerabilities in AI systems, including prompt injections 14:58 Roles of data controller and data processor under GDPR 19:47 Using S3 lifecycle policies for compliance and cost management 23:58 How to determine if a system is high-risk? 29:27 Human-in-the-loop, human-on-the-loop, and human-out-of-the-loop architectures 36:42 Embrace Cautiousness Strategically Join BLI University: https://discord.gg/jpJHGq6kgS Connect with Build Learn Impact: Instagram → https://www.instagram.com/build.learn.impact Substack → https://buildlearnimpact.substack.com Connect with Kevin Dowdy: Instagram → https://www.instagram.com/mrkevindowdy LinkedIn → https://www.linkedin.com/in/kevinldowdy Substack → https://substack.com/@kevinldowdy GitHub → https://github.com/kevindowdy DISCLAIMER: This is not professional advice. The views are my own and the people quoted. Consult your own advisers for legal, business or tax decisions based upon information from this episode. Build Learn Impact is on a mission to help our community create wealth and opportunity through technology. Subscribe if you’re ready to build the future.

  ListenShow Notes

Based on semantic analysis of episode topics and host coverage, this show is a strong guest fit for executives in:

Industry fit is computed by PitchCentric using vector embeddings of the show's episode catalog.