Build an AI Analyst with Claude Code in 50 Min | Sumeet Marwaha

Name: Build an AI Analyst with Claude Code in 50 Min | Sumeet Marwaha
Uploaded: 2026-01-18
Duration: 3107 s
Channel: Peter Yang
Description: Read a transcript preview of “Build an AI Analyst with Claude Code in 50 Min | Sumeet Marwaha” from Peter Yang, uploaded on 2026-01-18.

Channel: Peter Yang

Uploaded: 2026-01-18

Duration: PT3107S

Read a transcript preview of “Build an AI Analyst with Claude Code in 50 Min | Sumeet Marwaha” from Peter Yang, uploaded on 2026-01-18.

Transcript preview

Language: English (original, YouTube automatic captions)

[0:00] Eight months ago, it was just kind of

[0:02] debugging SQL. Now AI can really start

[0:04] to write that initial boilerplate SQL

[0:07] query and code that helps you then build

[0:09] on to do more advanced analyses. I've

[0:12] set those dashboards up many times in my

[0:14] career and they end up like kind of just

[0:15] getting ignored at some point. [music]

[0:17] Cloud will always read it. Cloud will

[0:19] always kind of start to ask questions.

[0:21] Recreating the way a data scientist

[0:23] actually thinks inside of a company.

[0:25] They're not just thinking of only the

[0:26] data. They're thinking of all the things

[0:27] outside of that. Podco is the first tool

[0:30] that I feel like could actually

[0:30] accomplish that.

[0:31] >> Which AI coding tool has the most

[0:33] momentum? Which one's gonna win?

[0:35] >> The customers on our platform have

[0:36] already picked it is definitely cursor.

[0:38] Cursor is absolutely kind of brushing it

[0:41] in startup coding tool of choice, but

[0:43] also enterprise coding tool of choice.

[0:46] >> So before we get into the cloud host

[0:48] stuff, can you talk about at a high

[0:49] level how AI can help with like typical

[0:51] data tasks? So yeah, here's kind of like

[0:54] the high level view of what I would say

[0:56] a typical analyst does in their their

[0:59] role and really how cloud code can start

[1:01] to augment that. You know, I could kind

[1:04] of think super high level and think

[1:05] about what kind of you would do from

[1:08] starting from scratch, but the reality

[1:09] is most people in their current job,

[1:11] they're not starting from scratch. They

[1:12] already have a series of dashboards.

[1:15] They have a series of queries they're

[1:16] running. And so at a