Person
1 episodes · 22 quoted claims
Chris Slovak says that about two to three years before the recording, many companies bought modern data warehouses such as Databricks or Snowflake.
“two, three years ago, everybody purchased a modern data warehouse, whether it's a Databricks or a Snowflake.”
Slovak says many companies are still in migration cycles of about two years to move data into centralized platforms.
“many, many companies are like, still like smack dab in these two year long cycles to migrate and import.”
Slovak says agentic systems can reason across multiple systems without one aggregated data source, provided they have context and semantic linking.
“they can reason like a human can across multiple systems, so long as there's context and semantic linking.”
Slovak says a retailer's program to move its data into Databricks and run a report would likely still be about twelve months away.
“But that program is probably still twelve months out.”
Slovak says he began his tech career working on SOAP APIs at Yahoo, which he describes as rigid and heavily documented.
“I got started in tech working on soap APIs at Yahoo, right? Super rigid, super documented, you know, APIs.”
Slovak says SOAP APIs gave way to the more flexible REST APIs, the programmatic web.
“And then that's given way to rest, which is a little bit more flexible, the programmatic web, if you will.”
Slovak says agent-to-agent communication and MCP-style protocols are rising, with agents asking other agents for information.
“You're seeing the rise of mCPS and agent to agent communications.”
Slovak says Unframe's inventory intelligence solution stores data in both relational and vector databases so it can return factual figures and also find similar items.
“We store the data in both, you know, relational as well as vector databases, so that we can do, you know, factual pull throughs.”
Slovak says Unframe's inventory intelligence deployment is estimated to save a customer between forty and fifty thousand dollars in time and data management costs.
“we're saving them somewhere between forty and fifty thousand dollars in both time and data management costs.”
Slovak says the inventory customer is projecting returns of forty times its investment.
“Not not only are they making great decisions and projecting out forty x ROIs,”
Slovak says a classic deep learning OCR model can extract information from a document page at a fraction of the cost of using a large language model.
“tell a classic deep learning OCR model? That's a fraction of the cost? Hey, look at this page and pull out the information that we need.”
Slovak says he has sold software-as-a-service solutions for about twenty years under a model of demonstrating features and leaving customers to build.
“I've been selling software solutions, SAS, for 20 years, and the model is pretty much let me show you all the features.”
Slovak says building software in-house is not cheap because maintenance and the talent needed are expensive.
“And everyone will tell you like, building is cheap, it is not at all, it's the maintenance that's expensive, right?”
Slovak refers to an outcome-based model in which Unframe is not paid until the solution works, as described by Larissa in an earlier episode.
“And you've heard Larissa talk about, we don't get paid until it works.”
Slovak says a multiline insurance company previously took about nine months to build one end-to-end use case.
“Well, in the prior world, it was taking them around nine months to build one full end to end, you know, system, use case solution.”
Slovak says the multiline insurance company now builds a use case with Unframe in two weeks.
“Now with us, they do it in two weeks.”
Slovak says nine-to-twelve-month projects costing millions of dollars cannot be allowed to fail, whereas two-week projects can.
“Nine to twelve month projects that are costing millions of dollars cannot fail. It's too much of an investment to fail.”
Slovak asks whether a system that is consistently about ten percent off is good enough to make a decision, and argues that usually it is.
“if the answer that you're getting from a system is consistent, but consistently ten off, is a ten gap good enough to make a decision?”
Slovak says that what others call ontology or context layers, Unframe calls a knowledge fabric, linking data, intent and tool usage.
“This is actually where technology context layers, as you know, Peter calls, like ontology layers, we call it knowledge fabric.”
Slovak says the aging boomer generation holds experience that is not written down, and that a context layer can help capture it for agents.
“the boomer generation is aging. They have tremendous wealth of experience and knowledge that, like, unfortunately, like, probably isn't written down somewhere.”
Slovak says that if a first use case covers about sixty of the major business entities and tools, the second use case is already largely covered.
“Use case number two is already like sixty of the way there.”
Slovak says he borrowed the idea of T-shaped versus I-shaped people from a former CEO, and uses it to describe agents that share context across departments.
“I call them T people. I stole this from my old CEO, so I got to give him a shout out.”