Hardware Podcasts

Edge Computing and AI: A Conversation with Ben Wolff

The Robot Report Podcast · 10 Oct 2025 · 1h 16m

AI in practice: one real use of AI described

What this episode covers

Ben Wolff, CEO of Palladyne AI (formerly Sarcos), discusses the company's pivot from hardware to a software-only model, focusing on Palladyne IQ for industrial robots and Palladyne Pilot for autonomous drone swarming. He explains the strategic advantages of edge computing for latency and security, the challenges of a 12-18 month sales cycle, and the company's commitment to building hardware-agnostic AI solutions that are indispensable to customers.

AI in practice

Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.

Engineering & simulationPalladyne AI

Palladyne AI developed domain-specific language models narrowly tailored to run on edge hardware (NVIDIA/Qualcomm boards) to enable autonomous drone swarming and robot programming without cloud connectivity.

“we've had to tear narrowly tailor our what we call domain specific language models. to be able to be constrained enough, from both a compute perspective, an energy consumption perspective, to be able to be done right there on the machine.”

Speaker 5 · Hear it at 58:31

Most relevant to

companies scaling up

Topics

Mentioned

Companies
Palladyne AI, Sarcos, SoftBank, ABB Robotics, Boston Dynamics, Figure, Parallax Systems, Red Cat, Teal Drones, FANUC, Zebra, Hyundai, Kuka, Yaskawa
Products
Atlas, Figure 3, Pepper
Chips
NVIDIA

Related episodes from other shows

Other shows on the same ground, closest first.

  • IoT For All Podcast · 17m

    The Future of Edge Computing and AI | Axelera AI's Fabrizio Del Maffeo | Internet of Things Podcast

    Fabrizio Del Maffeo, co-founder and CEO of Axelera AI, explains the shift from cloud-based AI to edge AI, emphasizing benefits like reduced latency, real-time decision-making, and enhanced privacy. He details how Axelera AI’s hardware accelerates computer vision and generative AI models on edge devices, addressing challenges in model compression and power efficiency. The conversation covers retrofitting existing infrastructure with AI, the complementary role of cloud computing, and industry trends observed at Super Computing 2024, including the rise of inference-focused hardware alternatives to Nvidia.

    AI & machine learningEmbedded firmware
    9 chapters

  • Over the Air · 33m

    Managing 250 Million Messages Per Day

    Eric Simone, Founder and CEO of ClearBlade, discusses the company's evolution from a services firm to a product-led Industrial IoT platform. He details the painful transition in 2013, including retooling staff and handling employee departures, and explains their strategic bet on edge computing before it was mainstream. Simone highlights ClearBlade's 'intelligent assets' platform, which allows non-technical operators to manage connected devices via no-code tools while remaining flexible for developers, and outlines their future focus on deploying AI models at the edge.

    Connected products & IoTLeadership, teams & culture
    7 chapters

  • IoT For All Podcast · 26m

    Edge Computing Benefits and Considerations | Pratexo's CEO Blaine Mathieu

    Blaine Mathieu, CEO of Pratexo, explains how edge computing unlocks the value of IoT and AI by processing data locally rather than relying solely on the cloud. He details use cases including bird monitoring at European wind farms and predictive maintenance for power transformers, highlighting benefits like latency reduction, privacy, and resilience. Mathieu also discusses distributed and swarm computing architectures that allow edge nodes to collaborate, ensuring high availability and scalability for industrial applications.

    Connected products & IoTAI & machine learning
    7 chapters

  • IoT For All Podcast · 30m

    Foghorn's Ramya Ravichandar | Ensuring Value with Edge AI in IIoT Applications

    Ramya Ravichandar, VP of Products at Foghorn, explains how the company uses edge AI to deliver real-time value in industrial and commercial IoT deployments. She details Foghorn's evolution from edge analytics to machine learning, highlighting use cases like defect detection at Stanley Black & Decker and workplace safety monitoring. Ravichandar argues that 'proof of value' is critical for scaling IoT projects, as edge computing enables low-latency responses and reduces bandwidth constraints compared to cloud-only architectures. She also discusses the complementary relationship between cloud-based model training and edge-based inference, and Foghorn's new 'Lightning' solutions for rapid deployment.

    AI & machine learningConnected products & IoT
    8 chapters

  • IoT For All Podcast · 20m

    What is Edge AI? | Barbara's David Purón | Internet of Things Podcast

    David Purón, CEO and co-founder of Barbara, explains that Edge AI combines edge computing with artificial intelligence to process data close to IoT devices, driven by the proliferation of 13 billion IoT devices, democratized AI tools, and powerful edge hardware like Jetson and Raspberry Pi. He details how industrial companies use Edge AI for use cases such as optimizing chemical usage in water treatment and detecting fraud in electric utilities, while noting that fragmentation of protocols and OT/IT cultural integration remain significant challenges. Purón positions Barbara’s platform at the 'far edge' to address cybersecurity and interoperability issues in unattended industrial environments, advocating for standards like OPC UA and MQTT to scale these solutions across enterprise infrastructure.

    AI & machine learningConnected products & IoT
    7 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by The Robot Report Podcast. The episode itself belongs to its publisher.