In this episode, we dive into how generative AI is shifting from experimentation to real product development—and why it’s no longer just for machine learning experts. Our guest is Hugo Bowne-Anderson, a well-known voice in the data and AI world. Hugo has spent years at the intersection of education, developer experience, and AI product thinking.

We explore how everyone—from software engineers and data scientists to product managers and designers—now has a seat at the table when it comes to building with AI. We also dig into why now is the time to move beyond “prompt and pray” and start creating applications people actually want to use.

Whether you’re hands-on technical or just getting started, this conversation will give you both insight and inspiration.

Key Topics Covered

• Why now is the right time for a broader range of people to start building with generative AI
• The shift from “prompt and pray” to real product development
• Empowering people beyond ML experts—PMs, designers, SWE, DS—to participate meaningfully
• Lowering barriers to entry with new tools (e.g. Cursor, Gemini, Replit)
• Building LLMs, Agentic AI, and RAG Systems: Chunking, JSON Outputs, Latency
• “Explorers and Settlers” Framework for AI Teams
• The importance of evaluation-driven development for iterating from prototypes to real-world apps
• How AI product development is becoming more collaborative and inclusive
• Final Thoughts: What to Build Next and Why You Should

Memorable Outtakes:

💬 We’re no longer in the ‘prompt and pray’ era. The bar is higher—but the opportunity is bigger too.”

💬 “ AI is not gonna replace you. People who do what you do using AI will replace you.”

💬 “Everyone can build with AI—not just engineers. PMs, designers, analysts… they all shape the product.”

🔗 References & Resources

- Hugo’s Upcoming course, Building LLM-Powered Software for Data Scientists and Software Engineers: https://maven.com/hugo-stefan/building-llm-apps-ds-and-swe-from-first-principles?promoCode=ODSC

- Use the code ODSC for an additional 10% off.

- Hugo’s LinkedIn: https://www.linkedin.com/in/hugo-bowne-anderson-045939a5/

- Hugo’s Twitter: https://x.com/hugobowne

- Hugo’s Substack Newsletter: https://hugobowne.substack.com/

- Systematic Survey of Prompt Engineering: https://arxiv.org/pdf/2406.06608

- O'Reilly Radar essay, Escaping POC Purgatory: Evaluation-Driven Development for AI Systems: https://www.oreilly.com/radar/escaping-poc-purgatory-evaluation-driven-development-for-ai-systems/

- Essay -  Beyond Prompt-and-Pray: Building Reliable LLM-Powered Software in an Agentic World: https://www.oreilly.com/radar/beyond-prompt-and-pray/

- Build Reliable LLM-Powered Apps with Evaluation-Driven Development, Hugo’s free 10-day email course: https://vanishing-gradients.kit.com/8fff155b0c

- The Vanishing Gradients podcast: https://vanishinggradients.fireside.fm/

- High Signal: https://high-signal.delphina.ai/

This episode was sponsored by:

🎤 ODSC East 2025 – The Leading AI Builders Conference – https://odsc.com/boston/

Join us from May 13th to 15th in Boston for hands-on workshops, training sessions, and cutting-edge AI talks covering generative AI, LLMOps, and AI-driven automation.

🔥 Use the exclusive code ODSCPodcast for an additional 10% off any ticket type!