Real-World AI Product Building - From 0 to scale with Saurabh Samtani

AI product building is easy to talk about.

Building an AI product that actually works for millions of users is a completely different game.

In this 75-minute masterclass, I'll take you behind the scenes of how real-world AI products are thought about, designed, built, launched and scaled - from a Product Manager's perspective.

This isn't another session on "What is GenAI?" or a list of AI tools.

We'll focus on the messy, practical parts that most PMs don't get to learn:

• How do you identify problems where AI actually creates meaningful value?
• When should you use an LLM vs traditional ML, search, RAG, rules or a combination?
• How do you design an AI experience beyond just putting a chatbot on the screen?
• How do you think about prompts, context, tools, agents and model selection as a PM?
• How do you handle hallucinations, latency, reliability and bad outputs?
• How do you evaluate the quality of an AI product?
• How do you balance quality × latency × cost?
• What changes when your AI product goes from a prototype to millions of users?
• How do you work with AI/ML engineers without needing to become an AI engineer?
• How do you experiment and iterate when AI outputs aren't deterministic?
• What does the actual AI product development lifecycle look like inside a large consumer tech company?

What makes this different?

I'll use real-world product examples and case studies to explain the decisions that happen behind the scenes — the trade-offs, failures, iterations and scaling challenges that rarely make it into AI tutorials.

By the end of the session, you'll have:

✓ A framework for identifying good AI product opportunities

✓ A practical mental model for designing AI products

✓ A PM-level understanding of the AI/LLM stack

✓ A framework for thinking about quality, cost, latency and scale

✓ A way to evaluate AI product performance

✓ A checklist for taking an AI feature from idea → prototype → production

✓ A much clearer understanding of what it actually takes to build AI products at scale

Who is this for?

PMs, APMs, aspiring PMs, founders and anyone who wants to move beyond experimenting with ChatGPT and actually learn how production-grade AI products are built.

No AI/ML engineering background required. :)