The Reasoning Show

SUMMARY: With the explosion of AI-generated code and applications, the modern SRE requires an AI-native approach to managing complex systems. 
GUEST: Anish Agarwal - CEO/Cofounder of Traversal
SHOW: 1016
SHOW TRANSCRIPT: The Reasoning Show #1016 Transcript
SHOW VIDEO: https://youtu.be/hF3MCRDhMno
SHOW SPONSORS:
Nasuni - Activate your data for AI and request a demo
ShareGate - ShareGate Protect. Microsoft 365 Governance, we got this!
SHOW NOTES:
Traversal (homepage)
Topic 1 - Welcome to the show. Tell us a little bit about your background, and what you focus on these days at Traversal. 
Topic 2 - AI is dramatically accelerating code generation, but not improving production outcomes. What’s fundamentally breaking in the traditional SRE model—and where do you see the biggest friction between speed and reliability?
Topic 3 - What are the most common failure patterns or mistakes you’re seeing in production from AI-generated code—and what’s driving them?
Topic 4 - AI can generate functional code, but it often lacks context about how systems behave in production. How is this changing what ‘good observability’ needs to look like?
Topic 5 - How do you see SRE evolving in an AI-first world? Does it become more automated, more policy-driven, or even partially autonomous?
Topic 6 - For organizations that want to embrace AI-assisted development but avoid production chaos, what are the most important guardrails they should put in place?
Topic 7 - If we fast-forward 2–3 years, what does a ‘modern’ production stack look like in a world where most code is AI-generated? What capabilities become absolutely essential? In one sentence—what’s the #1 thing a CTO should do right now?
FEEDBACK?
Email: show @ reasoning dot show
Bluesky: @reasoningshow.bsky.social
Twitter/X: @ReasoningShow
Instagram: @reasoningshow
TikTok: @reasoningshow

SUMMARY:  Today’s episode is all about a transformation happening in customer service—one that’s moving us from static systems and scripted workflows into something far more dynamic: AI systems that can actually learn and improve over time.
GUEST: Shashi Upadhyay (President of Product, Engineering, and AI at Zendesk)
SHOW: 1015
SHOW TRANSCRIPT: The Reasoning Show #1015 Transcript
SHOW VIDEO: https://youtu.be/IQaxE-DjIpo
SHOW SPONSORS:
ShareGate - ShareGate Protect. Microsoft 365 Governance, we got this!
Nasuni - Activate your data for AI and request a demo
SHOW NOTES:
The future of service belongs to self-improving AI
Topic 1 - Welcome to the show. Tell us a bit about your background and your focus today. 
Topic 2 - You describe this moment as a shift from systems of record to intelligent systems of action. What’s fundamentally broken in today’s customer service model that’s forcing this transition now? What changed in the last 2–3 years to make this possible?
Topic 3 - There’s been a lot of AI in customer service that overpromised and underdelivered. What are the biggest gaps between what customers actually need—like resolution—and what legacy automation has been delivering?
Topic 4 - The concept of a “self-improving” system is really powerful. What’s actually new here—what enables AI to improve with every interaction without constant human tuning?
Topic 5 - You’ve moved from assistive copilots to what you call “agentic AI” that can resolve issues end-to-end. Where are we today on that journey—and what still requires human involvement?
Topic 6 - Voice has historically been one of the hardest channels to automate. What changes with this new generation of AI that makes even complex, multi-step voice interactions solvable?
Topic 7 - If we fast-forward 2–3 years, what does a “best-in-class” customer service experience look like in an AI-first world?
FEEDBACK?
Email: show @ reasoning dot show
Bluesky: @reasoningshow.bsky.social
Twitter/X: @ReasoningShow
Instagram: @reasoningshow
TikTok: @reasoningshow

SUMMARY:  Brian (@bgracely) and Brandon Whichard (@bwhichard, Software Defined Talk and Failover Media) discuss the biggest AI news stories from the month of March, 2026. 
SHOW: 1014
SHOW TRANSCRIPT: The Reasoning Show #1014 Transcript
SHOW VIDEO: https://youtu.be/XwyAC-hxOQY
SHOW SPONSORS:
VENTION - Ready for expert developers who actually deliver?
Visit ventionteams.com
SHOW NOTES:
Links to all the AI News covered in this months show
FEEDBACK?
Email: show @ reasoning dot show
Bluesky: @reasoningshow.bsky.social
Twitter/X: @ReasoningShow
Instagram: @reasoningshow
TikTok: @reasoningshow

SUMMARY: With @bwhichard, we dig into how daily work-life changes when you make @AnthropicAI @claudeai the center of all workflow activities. 
SHOW: 1013
SHOW TRANSCRIPT: The Reasoning Show #1013 Transcript
SHOW VIDEO: https://youtu.be/zEmEH0t67js
SHOW SPONSORS:
VENTION - Ready for expert developers who actually deliver?
Visit ventionteams.com
SHOW NOTES:
Topic 1 - How long have you been living the Claude-life, and when did it dawn on you to make this central to your day-to-day activities? 
Topic 2 - What were the biggest hurdles you had to overcome before you trusted the system and started letting it have ownership over tasks and workflows?
Topic 3 - What are some of your best practices in terms of machine setup, how or where you store data, how you decide what to give it access to? Walk me through your thoughts around things like keeping things simple, where to be complex, how you think about security, etc.
Topic 4 - How are you learning to give it more responsibilities, or just figure out new ways to be productive with it? 
Good resources you’re pulling from? 
Any tips to make it use less tokens?
Skills marketplaces?
Topic 5 - What have been some of the biggest barriers to successful adoption, or just areas where you’re still struggling to get it to do the things you want? Or are you still in the learning curve stage and things just keep growing on one another?
Topic 6 - If you took the knowledge and skills you have now in Claude-life into your day-job, how do you see yourself working, as well as working with the rest of your team/teams? Would it bother you if you didn’t think they were using AI tools as much? 

FEEDBACK?
Email: show @ reasoning dot show
Bluesky: @reasoningshow.bsky.social
Twitter/X: @ReasoningShow
Instagram: @reasoningshow
TikTok: @reasoningshow

SUMMARY: We dig into the NVIDIA GTC keynote and highlight three things - accelerated computing for everything, the complexity of the new inference stack, and NVIDIA’s “open” software stack including NemoClaw.
SHOW: 1012
SHOW TRANSCRIPT: The Reasoning Show #1012 Transcript
SHOW VIDEO: https://youtu.be/aXOr91q76yM
SHOW SPONSORS:
VENTION - Ready for expert developers who actually deliver?
Visit ventionteams.com
SHOW NOTES:
NVIDIA GTC 2026 (Keynote)
NVIDIA NemoClaw - OpenClaw + OpenShell + NVIDIA Agent Toolkit
NVIDIA adds Groq LPU to their rack systems
NVIDIA to invest $26B in Open Weight Models
Interview with Jensen about Accelerated Computing (Stratechery)

Topic 1 - Jensen’s trying to paint the bigger picture of accelerated computing everywhere (robotics, autonomous driving, gen-ai, physical ai - but also just everyday enterprise apps). Everything is about keeping the stock price up, and margins high. The stock price provides the warchest to fight off all foes. 
Topic 2 - The inference architecture is a complex mix of GPUs, CPUs, ASICs/LPUs, high-speed networking and seems very different from the training architecture. How big is the burden on data center providers? What are the inference alternatives emerging? 
Topic 3 - Jensen talked a lot about OpenClaw and eventually about NVIDIA’s NemoClaw. How does his interest in Agentic AI tie into his interest in building NVIDIA’s own frontier model

FEEDBACK?
Email: show @ reasoning dot show
Bluesky: @reasoningshow.bsky.social
Twitter/X: @ReasoningShow
Instagram: @reasoningshow
TikTok: @reasoningshow