Generative AI in the Real World

AI makes it possible to generate code faster than teams can understand and support it. In this episode, Honeycomb cofounder and CTO Charity Majors joined Ben to discuss how that shift changes the work of building reliable software. Charity explained how observability helps teams connect changes with outcomes, why nondeterministic systems need “more rigor, not less,” and how traces can show the decisions agents make along the way. Over their wide-ranging conversation, she and Ben also covered automatic instrumentation with OpenTelemetry, why we should think of ourselves as “owners on the loop,” the limits of AI in debugging and SRE, and broader industry concerns like the pressure AI hype places on engineering teams and why maintenance, not speed, builds “enduring value.”

Pete Warden has spent his career on the frontier of small, local AI, first as one of deep learning's earliest engineers (he coined the term “TinyML”) and now as founder of Useful Sensors and Moonshine AI, where he builds voice models that run entirely on-device. Pete joined Ben to make the case that local AI no longer has to be a compromise. They get into what it actually takes to run a capable model on a laptop today; why the voice interface’s bad reputation is a consequence of rough, early implementations rather than a reflection of current capabilities; and where he stands in the ongoing debate between general “end-to-end” models and the compound AI approach of chaining specialized models together. Pete also explains why he thinks browser-based inference could be an "iPhone moment" for local AI and why more and more enterprises are considering self-hosted local models over commercial options. "The shape of [LLMs] is perfect for running locally," Pete says, and local models could be a boon to enterprises worried about cost, privacy, and stability.

A typical apartment building generates data on everything from leaking toilet flappers to tenant demographics, yet most of the real estate industry still runs its analysis by hand in spreadsheets. Fundrise co-founder and RealAI CEO Ben Miller argues the missing piece isn't a better AI model but a proprietary data layer that general-purpose tools can't replicate. Miller joined Ben Lorica on Generative AI in the Real World to cover how RealAI evolved from a data project into an AI-powered analyst tool, why the AI acts as an orchestrator rather than a financial calculator, which investment trends aren't yet pricing in AI's economic impact on real estate markets, and why he believes the trillion-dollar AI CapEx build is temporarily masking what will eventually be a deep trough in white-collar hiring.

The tech industry is measuring AI productivity all wrong, and Mozilla MLOps engineer and University of Chicago instructor Chelsea Troy makes a strong case for why. The real opportunity, she argues, isn't shipping more code faster but finally having the bandwidth to run the experiments, tests, and simulations that engineering teams have always wanted to run but never had time for. Chelsea joined Ben to cover the state of entry-level hiring, why the software engineering interview has been broken for decades, what it means to teach Python in 2026, and why token efficiency should replace token consumption as the industry's dominant productivity metric.

BERTopic creator and Google DeepMind developer relations engineer Maarten Grootendorst has spent years helping practitioners build intuition for how AI systems actually work—not just how to prompt them. Maarten joined Ben Lorica to cover the enduring relevance of embeddings and topic models in an LLM-dominated world, his hot take that agents are essentially just an “LLM in a for loop with some tools, some memory, and perhaps some guardrails," and what separates genuine agentic behavior from a well-constructed pipeline. They also get into the practical trade-offs between open weight and proprietary models, the future of state space models and attention, and why Maarten worries that a generation of builders shipping code they can't read may be storing up technical debt they can't repay. "If you don't really know how an LLM works," he says, "that intuition [about how to use it effectively] is much more difficult to develop."

As a pandas core contributor and early Parquet adopter who built AI data pipelines at streaming company Tubi TV, Chang She saw firsthand why the traditional data stack breaks down for AI workloads—and founded LanceDB to fix it. Chang joined Ben Lorica to explain why vector databases are too narrow a solution for modern AI data needs, and what a true multimodal data infrastructure actually looks like. Chang and Ben get into why the Lance file format is quickly becoming the open source standard for multimodal data, how the rise of agents is exploding data infrastructure demands, why open-weight models are the enterprise cost shift to watch in the next 12 months, and more. "Trillion is the new billion," Chang says, and the enterprises that set up their data infrastructure now for that scale will be the ones that succeed.

As the founder and CEO of LevelUp Labs, Aishwarya Naresh Reganti helps organizations “really grapple with AI,” and through her teaching, she guides individuals who are doing the same. Aishwarya joined Ben to share her experience as a forward-deployed expert supporting companies that are putting AI into production. Listen in to learn the value all roles—from data folks and developers to SMEs like marketers—bring to the table when launching products; how AI flips the 80-20 rule on its head; the problem with evals (or at least, the term “evals”); enterprise versus consumer use cases; and when humans need to be part of the loop. “LLMs are super powerful,” Aishwarya explains. “So I think you need to really identify where to use that power versus where humans should be making decisions.” Watch now.

Post-training gets your model to behave the way you want it to. As AMD VP of AI Sharon Zhou explains to Ben on this episode, the frontier labs are convinced, but the average developer is still figuring out how post-training works under the hood and why they should care. In their focused discussion, Sharon and Ben get into the process and trade-offs, techniques like supervised fine-tuning, reinforcement learning, in-context learning, and RAG, and why we still need post-training in the age of agents. (It’s how to get the agent to actually work.) Check it out.

Synthetic data has been around for a long time, decades even. But as KPMG’s Fabiana Clemente points out, “That doesn’t mean there aren’t a lot of misconceptions.” Fabiana sat down with Ben to clarify some of the current applications of synthetic data and new directions the field is taking—working with offshore teams when privacy controls just don’t allow you to share actual datasets, improving fraud detection, building simulation models of the physical world, enabling multi-agent architectures. The takeaway? Whether your data’s synthetic or from the real world, success often comes down to the processes you’ve established to build data solutions. Watch now.

SwirlAI founder Aurimas Griciūnas helps tech professionals transition into AI roles and works with organizations to create AI strategy and develop AI systems. Aurimas joins Ben to discuss the changes he’s seen over the past couple years with the rise of generative AI and where we’re headed with agents. Aurimas and Ben dive into some of the differences between ML-focused workloads and those implemented by AI engineers—particularly around LLMOps and agentic workflows—and explore some of the concerns animating agent systems and multi-agent systems. Along the way, they share some advice for keeping your talent pipeline moving and your skills sharp. Here’s a tip: Don’t dismiss junior engineers.