Applied Intuition blog

S2GO (Streaming Sparse Gaussian Occupancy) is a new approach to 3D scene understanding that builds a lightweight, streaming 3D occupancy map using a small set of learned 3D queries that evolve over time. It maintains a compact query-based world state, enabling dense, high-quality occupancy prediction up to 5.9× faster than prior methods and supporting long-horizon, camera-only, real-time perception. This blog post explains the motivation, method, and integration of S2GO into modern autonomy stacks.

SPACeR is a method for producing agents that mimic realistic, human-like traffic behavior at scale, trained using self-play reinforcement learning (RL). It enables fast, interactive simulation of many vehicles while keeping behaviors aligned with how people actually drive. This blog post will explain how SPACeR works, why it matters for AV simulation, and how it is integrated into Applied Intuition’s products.

In a fireside chat with Applied Intuition CEO Qasar Younis, JPMorgan Chase CEO Jamie Dimon described how curiosity, clear thinking, and strong management protect companies from decay. Among his warnings to leaders: that bureaucracy—not external threats—is often the silent killer of great institutions.

Autonomous vehicle simulation demands HD maps that scale. This post explores the technical evolution from monolithic files to cloud-native, tiled architectures. Learn how distributed services, parallel ingestion, and incremental migration practices doubled simulation throughput while cutting infrastructure costs. Real-world results show how scalable HD mapping powers the future of autonomy testing and safety validation.

From an expanded base in Stuttgart-Vaihingen, Applied Intuition is rolling out more Self-Driving System programs in Europe. The enlarged hub features garages and workshops for fleet operations and a growing team building ADAS and automated driving features, leveraging a white-box autonomy stack that shortens time to market for European automakers.

Applied Intuition’s 2025 was the year vehicle intelligence moved from vision to reality. This look back highlights flagship product launches, high-impact partnerships in automotive, trucking, mining, and defense, rapid global team growth and new offices, high-profile thought leadership, and a landmark Series F financing at a $15B valuation.

Ensuring dependable ADAS simulation and log replay remains a technical challenge, especially when run across varying hardware and cloud environments. This post examines how Action Graph addresses these issues, enabling more reliable validation, repeatability, and performance for autonomous systems.

Peter Ludwig, Applied Intuition Co-founder and CTO, testified before a House committee on China’s auto threat, physical AI, and software-defined vehicles. His testimony explains how cars are becoming strategic infrastructure, how dual-use autonomy links industry and defense, and where Chinese-connected vehicles create economic and security challenges the United States must actively address.

Explore how Ford Otosan leveraged Applied Intuition’s simulation platform and TruckSim models to advance ADAS development. The collaboration demonstrates a rigorous, data-driven validation process, fostering safety, efficiency, and trust in virtual vehicle simulations

Understanding ODD coverage helps AV developers spot untested conditions, fill data gaps, and justify safety readiness. Learn how Applied Intuition’s Validation Toolset supports smarter coverage analysis through simulation, data tagging, and statistical confidence methods.