Dana for Physical AI

Dana is Applied Intuition's agentic platform for physical AI—a new way to build physical AI, made for the complexities of building and deploying intelligent, safety-critical machines. It brings a decade of Applied Intuition's infrastructure, tooling, and data expertise into one platform for the data and sim flywheel behind autonomous vehicles, trucks, and robots. Building the autonomy flywheel—data → sim / train / eval → insights—is incredibly hard, takes deep domain expertise and years to get right. Dana brings a decade of Applied Intuition's infra, tooling, and data expertise into a single agentic platform to solve autonomy's hardest problems: 🔷 Curate datasets from petabytes of fleet data, across every domain 🔷 Reconstruct scenes with Gaussian splatting and run neural simulations 🔷 Augment behaviors with RL-trained agents and generate diverse sensor data with world models 🔷 Query results and build dashboards that turn every result into action Our PM, Gautham Sholingar, gives a peek inside. Explore Dana: https://applied.co/dana-blog?utm_source=youtube_com&utm_medium=organic&utm_campaign=dana — What is Dana? Dana is Applied Intuition's agentic platform for physical AI. It replaces fragmented tools and manual handoffs with agentic workflows, unifying data curation, simulation, training, evaluation, and insights in one platform you operate through natural language, APIs, and SDKs. What can Dana do? Dana curates datasets from petabytes of fleet data across every domain, runs inference and neural simulation, reconstructs scenes with Gaussian splatting, generates adversarial behaviors with RL-trained agents and world models, and compares autonomy stacks with full traceability in a central metrics lakehouse. Who is Dana for? Teams building physical AI and autonomy—autonomous vehicles, software-defined vehicles, trucks, and robots—that need a safety-critical, traceable development flywheel. Why does it matter? Dana is part of Applied Intuition's mission to make a billion machines intelligent, turning workflows that once took months into days—and, in some cases, hours. Chapters: 0:00 A new era of intelligent, autonomous machines 0:09 The autonomy flywheel—and why it's so hard to build 0:35 Introducing Dana 0:43 A peek inside, with Gautham Sholingar 0:55 The three steps of every flywheel stage 1:10 Data: curate from petabytes of fleet data 1:48 Workflows: sim, train, eval & inference at scale 2:01 Gaussian splatting & neural simulation 2:19 Edge cases: RL-trained agents & world models 2:44 Insights: dashboards that turn results into action 3:21 One agentic platform—that's Dana