Scaler Pod - The Software Engineer's Podcast · Scaler Academy

GATE AIR-2 Srikanth Varma on Data Science, Job Roles & Future Scope | SCALER POD 11

·1 hr 23 min·5 clips

He argues the best model is often a simple baseline, not the fanciest neural network.

Varma joins Scalar Pod as someone very close to home. He starts with a plain account of what his day looks like now. That day is split between Applied Roots and Scalar's data science and machine learning work. He treats both sides as serious responsibilities rather than side projects. The first priority is keeping scheduled classes on track. Instruction stays at the center of the discussion. He says that in EdTech, instruction is the core. He also spends regular time with instructors to make sure learners are getting the best possible experience. Hiring takes up a large share of attention. The bar is high, so bringing in new people is never simple. When instructors join, they may already be terrific scientists, but the role still needs mentoring. The conversation keeps coming back to the idea that practical teaching and team support matter as much as technical depth. From there, the episode shifts into a long, useful example about modeling. Varma talks through a case where someone built a fancy prediction system without first checking a baseline. He explains that the basic comparison should come first. A simple Python script that averages numbers can already tell you a lot. Even without advanced math, a smart beginner can build something that sets a floor for judgment. The point is not to reject machine learning. The point is to avoid pretending that sophistication alone proves value. He stresses that any model should be measured against a common-sense reference. If the fancy result is not better than the baseline, the work has not yet earned its complexity. That idea becomes one of the episode's clearest lessons. It also reflects the show's general style, which stays conversational and respectful while moving through work history, role design, and practical learning. The episode closes with appreciation for the conversation and interest in bringing more data science and machine learning guests onto the show.

As heard by us

A practical look at building a data science program, from hiring and mentoring to the value of a simple baseline.

ScalerPod centers on the practical work of building a data science program. Varma, founder of Applied Roots and now leading Scalar's data science and machine learning vertical, splits his time between both efforts.

Read the full review in PlayNext →

Why you'd press play

You want a plainspoken look at how Scalar builds its data science program.

Read the full recommendation in PlayNext →

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from Scaler Pod - The Software Engineer's Podcast