Hosted by Making Data Simple
Hosted by Al Martin, WW VP of Technical Sales at IBM, Making Data Simple cuts through the hype to reveal how data and AI reshape the modern enterprise.
405 episodes · publishes weekly · latest 2026-07-01 · ~41 min/episode
Rank
#42
Substance
86.0
/ 100
Breakdown
Scored 2026-07
Updated monthly
General rank
#3 of 267
Across the index
#42 of 6182
Substance
Top 1%
outscores 99% of the index
Making Data Simple ranks #42 on The B2B Podcast Index with a substance score of 86.0 out of 100, scored across 1 recent episode. It scores highest on guest caliber and insight density. The guest literally co-invented RAG at Meta FAIR, was Head of AI Research at Hugging Face, is a Stanford adjunct, and now runs an enterprise RAG company serving Fortune 500 clients - about as relevant and senior as it gets for this topic.
Averaged across 1 recently scored episode, with cited evidence.
Several non-obvious ideas surface - fine-tuning elicits rather than injects knowledge, RAG as a compound system of ~12 models, 'make AI good enough for your data' vs cleaning data, and observability mattering more than accuracy - but they're diluted by extended product pitching and go-to-market Q&A.
“injecting new knowledge into a base model like that um, is just not possible”
“we need to make the AI just good enough for your data, however terrible your data is”
The 'false dichotomies' framing, the inversion of 'get data ready for AI' into 'get AI ready for your data', and the claim that observability trumps accuracy are genuinely fresh takes from a credible source, though much of the RAG-vs-fine-tuning discussion is well-trodden.
“don't believe in false dichotomies”
“You should probably just wait for AI to be good enough to work on any type of data”
The guest literally co-invented RAG at Meta FAIR, was Head of AI Research at Hugging Face, is a Stanford adjunct, and now runs an enterprise RAG company serving Fortune 500 clients - about as relevant and senior as it gets for this topic.
“led the research team at Meta that introduced RAG research way back in 2020. He also has been the head of AI research at Hugging Face”
“I spent five years at fair, as you said, um, um, doing a bunch of different research projects including RAG retrieval, augmented generation”
Some concrete anchors appear - Qualcomm's millions of technical PDFs, ~12 models per pipeline, dollars-per-token pricing, 70-80% vs 90%+ accuracy thresholds - but many claims stay abstract, with few hard metrics, timelines, or dollar figures.
“they are comprised of around 12 different models and one of them is the language model”
“what we're doing with Qualcomm where they have these extremely complex technical questions”
The host asks a few sharpening follow-ups ('double click on best in the world', 'methodology or the tech?') but largely lobs softballs, enables an extended sales pitch, agrees rather than pushes back, and self-identifies as 'a terrible host' before drifting to 'what do you do for fun'.
“I'm trying to get you to double click on best in the world specialized RAG agents”
“But is it a methodology or is it the tech underneath or is it both?”
First period on the Index - history builds from here.
1 scored on substance · 60 tracked in total.
Add this badge to your site - it links back here and updates automatically as you rank.
<a href="https://index.fame.so/show/making-data-simple" target="_blank" rel="noopener">
<img src="https://index.fame.so/badge/making-data-simple/badge.svg" alt="Ranked #3 on The B2B Podcast Index" width="360" height="136" />
</a>Track Making Data Simple's rank
Get an email whenever this show moves up or down the Index. Monthly at most, no spam.
Companies, products and tools that come up most across this show's episodes.
The themes that come up most across this show's episodes.
Podcasts that dig into the same topics.
