Connect your Claude, Codex or OpenClaw for full transcripts on millions of podcasts. Two week free trial, cancel anytime.

Start free trial →
DS Dharmesh Shah Dharmesh Shah Co-founder and CTO of HubSpot, and creator of Agent.ai, a network for AI agents, and WordPlay.com.

“Anything that can be solved with a structured query deterministically, solve with a structured query deterministically. Don't use inference to solve something that a query can solve… if your question is, how many customers or how much revenue are we getting from Ireland? Well, you shouldn't have to trawl through 8 million Slack messages and emails or whatever, infer the answer. … That should be a database query. That should not be solved through inference and throw a 300 IQ model at it.”

a16z Podcast · Startups & Venture · November 2026

“Anything that can be solved with a structured query deterministically, solve with a structured query deterministically. Don't use inference to solve something that a query can solve… if your question is, how many customers or how much revenue are we getting from Ireland? Well, you shouldn't have to trawl through 8 million Slack messages and emails or whatever, infer the answer. … That should be a database query. That should not be solved through inference and throw a 300 IQ model at it.” — Dharmesh Shah, a16z Podcast

Context

HubSpot co-founder Dharmesh Shah argues that business software should pick the cheapest, lowest-latency way to deliver an outcome. This is his rule of thumb for when not to use an LLM.

Transcript

a16z Podcast Around 22:53 into the episode
Dharmesh Shah

…of great commercial products. So as long as you stay off the yellow brick road, it is completely possible now that you're not going to get margined out or whatever because you have options. There's not just one company providing the one key ingredient to building a. An AI, an AI-native company. No,

Joe Schmidt

…building some sort of increasing return to scale or compounding advantage at this layer. And so I'd love to hear how you think about this at HubSpot or any of the things you're working on as it relates to the loops you try to build in these products or the advantages you're trying to build. And

Dharmesh Shah

it's compounding loops all the way down, Joe. Of course. That's the entire name of the game before AI, after AI. But a couple of things I'll say. So I'm going to rewind. I pushed one idea on the stack. I'm going to pop it off and talk through it. Is that so? One of the questions we get back on the, if you're an entrepreneur kind of evaluating the ideas, it's like, okay, well, so today was OpenAI's dev day, by the way, September 30th. So right before this call is data recording. You're recording it. What's that? I

Joe Schmidt

said we're dating the recording for the audience. So you now know timestamped.

Dharmesh Shah

Exactly. You know exactly when this is being recorded. So HubSpot launched our, we were the first CRM company to do so our connector to ChatGPT over a year ago, and we have similar one for Anthropic. And one of the questions we get sometimes is like, oh, well, doesn't that sort of diminish the…

Joe Schmidt

And I think there's many companies that are navigating through this as they try to understand their own pricing and consumption ideas here, which could be its own dang podcast, I think, for that matter. Maybe last question on this, because I think it's really interesting you talk through this. The…

Speaker names come from our own diarization. The position is estimated from where the line sits in the episode.

More from a16z Podcast