Sourish Samanta, Director AI and ML at Advance Auto Parts, joins The Tech Trek for a grounded conversation on where machine learning still creates the most business value, where generative AI fits, and why many teams are chasing the wrong solution. This episode is worth your time if you want a clearer view of how serious operators think about AI strategy, product delivery, and practical use cases that can ship now.

This conversation cuts through the noise around AI and gets back to first principles. Sourish explains why machine learning remains the foundation behind today’s AI wave, how to choose between deterministic and creative systems, and what it actually takes to build production ready products that solve real business problems.

In this episode:

Why machine learning is still the core layer behind modern AI

When to use machine learning, when to use generative AI, and when simple analytics is enough

What a real product mindset looks like for AI and ML teams

How pod based teams can ship faster with better cross functional alignment

Why AI and ML talent need to spend time continuously reskilling

Timestamped highlights:

00:00 Why machine learning remains the foundation of today’s AI stack

01:57 The difference between ML teams, AI teams, and agent focused workflows

05:56 Choosing the right solve, from forecasting and inventory to creative content generation

10:09 The product mindset required to turn AI ideas into working systems

13:51 Why some business problems need analytics, not AI

15:52 Why AI teams need to spend part of their time learning, testing, and staying current

Standout line:

AI is not the strategy. Solving the right problem is.

Practical takeaway:

If you are leading an AI initiative, start by classifying the problem. If the outcome needs consistency, prediction, or forecasting, machine learning may be the better path. If the outcome needs creativity or flexible generation, generative AI may be a better fit. And in some cases, the best answer is still a clean dashboard and strong analytics.

Follow The Tech Trek for more conversations on AI, data, engineering, and how technology actually gets applied inside real businesses.

Machine Learning: What Businesses Might Actually Need