Why 95% of Enterprise AI Projects FAIL! CEO of Apollo GraphQL, Matt DeBergalis
The Programming PodcastShow Notes
AI is writing more code than anyone expected. Some of it is great. A lot of it is just okay. In this episode, Danny Thompson and Leon sit down with Matt DeBergalis, CEO of Apollo GraphQL, to unpack what it will take to move from a gold rush of mediocrity to production-grade agentic experiences that users can trust.
Guest Co-Host: Matt DeBergalis https://www.linkedin.com/in/debergalis/ https://www.apollographql.com/ @ApolloGraphQL
SITE https://www.programmingpodcast.com/
Stay in Touch: 📧 Have ideas or questions for the show? Or are you a business that wants to talk business? Email us at [email protected]!
Danny Thompson https://x.com/DThompsonDev https://www.linkedin.com/in/DThompsonDev www.DThompsonDev.com
Leon Noel https://x.com/leonnoel https://www.linkedin.com/in/leonnoel/ https://100devs.org/
📧 Have ideas or questions for the show? Or are you a business that wants to talk business? Email us at [email protected]!
We dig into the real gap behind AI project failures and it is not the models. Matt explains why agentic development stalls inside enterprises, how microservice sprawl blocks useful AI, and where GraphQL functions as the control plane that unifies data, streaming, and context so agents can actually do work. We cover the early hype around MCP servers, why many of them ship without OAuth, and a concrete checklist for securing costs and credentials before you flip the switch.
You will hear where shopping, search, and SEO are headed as prompt boxes replace search boxes. We get into the gravity that pulls models toward stacks with the most public code, what that means for React, Rust, Python, and the long tail, and how developers can future proof their careers by mastering fundamentals like orchestration, context control, and system design instead of chasing every weekly model benchmark.
We wrap with a practical path for job seekers. Breadth over tool loyalty. Weekly small projects. Use AI for the first 75 percent, then own the last 25 percent with clear prompts and better workflows.
Who this episode is for - Engineering leaders trying to turn AI prototypes into products - Senior and staff engineers learning agent orchestration - Devs curious about MCP, GraphOS, and secure tool calling
You will learn - Why 95 percent of agentic projects fail and what capability is missing - How GraphQL unifies fragmented systems for agents, including streaming and precise context selection - A security and cost control checklist for MCP style tool calling - How hiring rubrics are shifting toward communication, systems thinking, and curiosity - A weekly practice plan to build portfolio proof fast
Highlights
Gold rush of mediocrity and what to do about it
From REST to stateful agents and why the old web stack creaks
Every search box becomes a prompt box
The 75 and 25 rule for productive AI assisted coding
Tool breadth over tool loyalty for career advantage
Chapters
00:00 Cold open. Why most agentic projects fail 01:00 Theme setup. The gold rush of mediocrity 01:30 Host and guest introductions 03:00 MCP excitement vs reality. From laptop tools to real products 06:15 Security and spend. OAuth gaps, scoped keys, rate limits, audit logs 09:00 Distribution shift. Generative SEO and agentic checkout 13:10 Centralization gravity. Why models favor stacks with more public code 18:00 Foundations. Unifying services with GraphQL and streaming tokens 24:10 Controlling the context window with field selection 26:30 Should developers learn this now 31:30 Fundamentals over benchmarks. MCP, RAG, evals 42:00 Hiring in the agent era. Communication, systems thinking, curiosity 48:00 Prompt quality and the last mile 53:00 Audience question. Tools to explore and a weekly practice plan 59:30 Closing recap and CTA
More from The Programming Podcast
Learn how to become a better software engineer from an Apple veteran with over 12 years of experience building products.
Anthropic's IPO is leaked and OpenAI DevDay 2026 just dropped a $500 AI plan, Decisions API, cloud-powered Codex, Dots, and a much bigger vision for AI agents. At the same time, Anthropic is reportedly targeting a massive IPO while the economics of frontier AI are getting harder to ignore.So what actually matters for developers?🤝 Thank You to Our Sponsors!GitKraken: Stop fighting Git and spend more time building. Get visual Git workflows, PR management, and tools built for modern AI-powered development. Check out exclusive offers at https://www.gitkraken.com/programmingpodLevel Up Financial Planning: Take your financial confidence to the next level with certified financial planner Lucas Casares, specializing in early and mid-career tech professionals. Get a complimentary meeting at https://www.levelupfinancialplanning.com/🔗 Resources & Links MentionedUndersold - https://www.undersold.ai/workspaceThe Developer's Guide to AI: Grab the newly released book by Danny Thompson from here! http://developersguide.ai/Stay in Touch:📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]!Danny Thompsonhttps://x.com/DThompsonDevhttps://www.linkedin.com/in/DThompsonDevwww.DThompsonDev.comLeon Noelhttps://x.com/leonnoelhttps://www.linkedin.com/in/leonnoel/https://100devs.org/📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]!In this episode of The Programming Podcast, Danny Thompson and Leon Noel break down the biggest AI and developer news of the week from the perspective of people actually using these tools every day.We get into:• Anthropic’s IPO, explosive revenue growth, infrastructure costs, and what the economics of frontier AI could mean for developers• OpenAI DevDay 2026 and the announcements we think actually matter• The new $500 ChatGPT plan and who could realistically justify paying for it• GPT-6.1 Sol and OpenAI’s push toward faster, higher-volume AI workflows• Decisions API vs. Jev and why decision-focused models are becoming interesting• Codex moving into the cloud and why cloud agents could change how developers work• Dots, Grokbot, Muse, and the shift from chatbots to always-on AI assistants• Why open-weight models are taking a larger share of real-world AI usage• Whether Notion, Slack, and traditional SaaS tools should be worried about AI-native workspaces• How Danny and Leon are actually using AI agents in their day-to-day workflows• Real Grokbot use cases for email, research, QA, content, development, and delegation• The difference between asking AI for answers and giving AI ongoing responsibilitiesThe bigger conversation underneath all of this:AI is moving from something you prompt to something you delegate to.And that changes what developers need to learn next.
Ranking the latest AI models from S tier to F tier. We break down when to use each one and which ones to avoid for your workflow. 🤝 Thank You to Our Sponsors! GitKraken: Stop fighting Git and spend more time building. Get visual Git workflows, PR management, and tools built for modern AI-powered development. Check out exclusive offers at https://www.gitkraken.com/programmingpodLevel Up Financial Planning: Take your financial confidence to the next level with certified financial planner Lucas Casares, specializing in early and mid-career tech professionals. Get a complimentary meeting at https://www.levelupfinancialplanning.com/ 🔗 Resources & Links MentionedThe Developer's Guide to AI: Grab the newly released book by Danny Thompson from here! http://developersguide.ai/ Stay in Touch:📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]!Danny Thompsonhttps://x.com/DThompsonDevhttps://www.linkedin.com/in/DThompsonDevwww.DThompsonDev.comLeon Noelhttps://x.com/leonnoelhttps://www.linkedin.com/in/leonnoel/https://100devs.org/📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]! Danny and I put popular tools to the test to see what actually holds up under pressure. We go through our own specific AI model ranking, debating the merits of each based on real-world application rather than just marketing hype. Since we did not compare notes beforehand, you get our honest, unfiltered thoughts on which systems are worth your time and which ones are currently failing to deliver. This LLM comparison covers the nuance of daily usage, helping you decide where to invest your focus. We also discuss how to optimize your AI usage to stay efficient without getting distracted by every new release. Whether you are a developer looking for better productivity or just trying to navigate the crowded landscape of tech tools, this breakdown gives you the straight answers you need to make better decisions. Subscribe for weekly software breakdowns, and let us know in the comments which model you think deserves the top spot.
Stop losing job offers due to poor interview preparation. Learn why a three-minute interview cutoff signals a lack of care.Most candidates assume that getting the interview is the hard part, but showing up unprepared is a fatal error. This breakdown examines a real-world scenario where a recruiter ended a conversation after only three minutes because the applicant couldn't demonstrate basic knowledge about the company. When you neglect to perform 30 seconds of homework, you communicate that the position is just another line on your resume rather than a priority.Proper interview preparation requires more than just rehearsing your own bio. It demands that you frame your value through the lens of what the employer actually needs. By failing to research the organization, you signal to the hiring manager that your interest is surface-level at best. Adjusting your strategy to include this quick research step ensures you stay in the conversation longer and separates you from the crowd.🤝 Thank You to Our Sponsors!GitKraken: Stop fighting Git and spend more time building. Get visual Git workflows, PR management, and tools built for modern AI-powered development. Check out exclusive offers at https://www.gitkraken.com/programmingpodLevel Up Financial Planning: Take your financial confidence to the next level with certified financial planner Lucas Casares, specializing in early and mid-career tech professionals. Get a complimentary meeting at https://www.levelupfinancialplanning.com/🔗 Resources & Links MentionedThe Developer's Guide to AI: Grab the newly released book by Danny Thompson from here! http://developersguide.ai/Commit Your Code Conference (CYC 2026): Join us Sept 2nd-4th at the Capital One Campus in Plano, Texas. Keynote by the CTO of the PGA! https://www.commityourcode.com/Resilient Coders: Free coding boot camp paying you a stipend to learn. Applications open in Boston, Philly, NY, Maine, and Pittsburgh: https://www.resilientcoders.org/Plus, we cover the Friday 3 PM developer log habit, why you should always be interview-ready every six months, and answer a listener question about how to handle recruiters before your first interview round.✅ If you made it to the end, comment: “Better talking points!” so we know who is a real one!https://www.commityourcode.com/Come to CYC!Get The Developers Guide To AI here!http://developersguide.ai/Stay in Touch:📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]!Danny Thompsonhttps://x.com/DThompsonDevhttps://www.linkedin.com/in/DThompsonDevwww.DThompsonDev.comLeon Noelhttps://x.com/leonnoelhttps://www.linkedin.com/in/leonnoel/https://100devs.org/📧 Have ideas or questions for the show? Or are you a business that wants to talk business?Email us at [email protected]!
Is Opus 5 worth the hype? I spent two days testing the 2026 release to see if the performance lives up to expectations.This Opus 5 review provides an honest look at the latest model, moving past marketing claims to examine real-world usability. If you are deciding whether to upgrade or switch workflows, this analysis breaks down the actual capability of the software compared to previous versions.I cover the specific Opus 5 benchmarks I recorded during my testing phase, highlighting where it succeeds and where it falls short. You will see my candid Opus 5 impressions after two days of heavy use, helping you determine if this model fits your specific needs or if you should wait for further updates.Subscribe for weekly software breakdowns, and let me know in the comments if you want to see a head-to-head comparison between Opus 5 and other top-tier models.