Spotted 👀

The result: → On semantic segmentation, LFG surpasses its own teacher on future frames (0.751 vs 0.680 mIoU), despite predicting blind while the teacher saw the real images. → Depth predictions on future frames closely match the teacher (0.31 vs 0.26 AbsRel on KITTI-360, 0.22 vs 0.19 on Waymo Open Dataset). → It correctly separates moving objects from static background even when the automated labeling pipeline gets it wrong.

Architecturally, the model consists of a frozen image encoder backbone paired with trainable causal autoregressive transformers. To reconstruct scenes and generate future predictions, these transformer layers are trained under the guidance of a suite of teacher models for depth, poses, semantic and motion.

When you're building a top tier team aimed at solving the most complex problems in physical AI, mediocrity is not an option. Over the years, @malharhar has interviewed and hired many across Applied. As a high growth company, one of the biggest challenges has been maintaining and evolving the culture at scale. review.firstround.com/firsth…

Hundreds of drones, built by different vendors, tasked with one mission. Before they deploy, someone has to check the pieces work together, and today that usually happens at a live event, often the first time these systems actually talk to each other. Applied Intuition's Digital Proving Grounds fix that — an always-on, vendor-neutral sim environment spanning the full sim-to-live continuum: 🔷 SIL: software tested continuously in a shared DevOps environment 🔷 HIL: real hardware verified against simulated environments 🔷 LVC: live platforms stress-test C2 at scale, no full fleet required 🔷 Live exercises: validate readiness, don't discover gaps Applied Intuition partnered with the @DoWCTO on Virtual Readiness and Experimentation, integrating air, surface, and subsurface vehicles from different vendors on one mission, compressing the find-fix-fly loop from weeks to under an hour. Now scaling across the @CDAODoW's Autonomy Factory. appliedintuition.com/blog/di…

We're headed to @UCBerkeley tomorrow, September 15th, for the Cal Engineering & Technology Career Fair. Come by booth #21 to learn more about the physical AI problems we're solving at Applied, from autonomous vehicles and agriculture to trucking, defense, and beyond. appliedintuition.com/careers…

Great code is only half the job. Making it work reliably on physical hardware in the real world is the other half, and it's the part most people don't get to touch. The ability to drive impact and take ownership has been a big reason why our Deputy CTO, @malharhar, has stayed at Applied all these years. He has been able to go where the problems are. Read his @firstround write up to learn more about his experience: review.firstround.com/firsth…

At this year’s Agentic AI Summit hosted by @BerkeleyRDI, our Chief Scientist, @Wei_ZHAN_ shared some of the cutting edge research he has been leading in the physical AI space. As the industry races to deploy L2++ ADAS with imitation-learning-based E2E, Wei tackled a key question: what if end-to-end autonomy could be trained without imitation at all, relying only on reinforcement learning? An idea that runs against the current status quo. Here are some of his insights 🧵

On the road. We’re deploying physical AI across Saudi Arabia, starting with driverless trucks. Our collaboration with @HUMAIN is laying the groundwork for intelligence at national scale. appliedintuition.com/blog/ap…

Think big. That’s always been a driving motivator for us, and now we have a partner to match our ambition. Today, we’re proud to announce Applied Intuition’s strategic collaboration with HUMAIN to deploy physical AI across Saudi Arabia, starting with autonomous trucking. This is a major step toward our goal of making a billion machines intelligent. Together, we will deploy thousands of autonomous trucks throughout key Saudi logistics corridors by 2030. This will make it the largest autonomous trucking network in the world, and our Self-Driving System and Vehicle OS technology will be the centerpiece of this collaboration with HUMAIN. This is physical AI at national scale. appliedintuition.com/blog/ap…

40 Engineers. 3 Days. One mission: bring physical AI to flight. During our recent hackathon, our defense team accelerated deployment pipelines, ran full HIL/SIL dress rehearsals, and integrated collaborative behaviors to edge compute. Drone software, air combat software, and hardware teams worked side by side to get there. Want to see your code fly? We're hiring: appliedintuition.com/careers…

With 1,000+ engineers on our team, it's no surprise the average person at Applied Intuition either comes from a technical background or enjoys teaching themselves technical things. Join us: appliedintuition.com/careers…

Applied Intuition has earned the VIVALDI Inspection Certification from TÜV SÜD, independently verifying our platform for virtual validation of automated driving systems. Under UNR171, it is not only about proving the AV system works but also proving the simulation tools are trustworthy. TÜV SÜD evaluated more than 100 requirements across key areas: 🔷 Scenario generation and coverage 🔷 Test management and reporting 🔷 Data integrity and auditability 🔷 Simulation determinism and scalability Our platform met or exceeded VIVALDI standards across every category. appliedintuition.com/blog/tu…

Dana. AI for the real world.

Dana for software-defined vehicles brings the speed of AI to development workflows. We built Dana to reimagine how physical AI applications get developed, including the vehicle software development lifecycle—adding an agentic layer so engineers can build, code, and ship intelligent vehicle features faster and at scale. appliedintuition.com/blog/fu…

Most autonomy pretraining still depends on data like lidar scans, HD maps, and hand-annotated trajectories and labels. Our research team challenged that norm and asked: what if a model learned to drive by watching dashcam footage instead? LFG, Learning to Drive is a Free Gift, was developed to do just that. The entire system is trained on roughly two million clips from freely available driving videos, spanning varied roads, conditions, and traffic situations. 85.2 PDMS on NAVSIM. One camera. No lidar. No maps. Read the full breakdown: appliedintuition.com/researc…

Dana keeps the flywheel moving and the system learning with every turn. appliedintuition.com/blog/da…

Applied Intuition and @aerovironment have announced a strategic teaming agreement to advance uncrewed teaming capabilities for the U.S. military and its allies. Applied Intuition's Acuity ISR/Strike software will integrate into AV's Mayhem 10™ launched effects system, enabling teams of Mayhem 10 to autonomously find, fix, track, target, and engage targets at the tactical edge — under the supervision of a single operator. The collaboration supports the U.S. Army's Launched Effects program and is designed to scale across short, medium, and long range requirements. appliedintuition.com/press-r…

Engineering has changed. So has our interview process. appliedintuition.com/enginee…

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 ⬇️

Introducing Dana. Agentic AI for the physical world. appliedintuition.com/blog/da…

Most companies build the aircraft, then integrate autonomy specific to that airframe. Here at Applied Intuition, we took the opposite approach. Drone Stack is hardware-agnostic from day one, running on any platform with sufficient compute. appliedintuition.com/blog/dr…

"The upside is there's more opportunity." Carolyn Davis, U.S. Automotive Lead here at Applied Intuition, gives her take on driving impact at a high growth company. appliedintuition.com/careers…

World foundation models can generate sensor data for scenes you've never driven. Any weather, any lighting, any ODD. World foundation models like @nvidia Cosmos unlock new sensor data capabilities. Making them production-ready takes a complete toolchain. Applied Intuition built a complete toolchain so autonomy developers can turn fleet data into diverse, validated sensor datasets for production use. We built a five-stage reference pipeline with NVIDIA Cosmos: ➡️ Curate + auto-label fleet data into conditioning-ready segments ➡️ Extract scenario, map, and sensor geometry to ground the model ➡️ Generate recipes, conditioning, and prompts to guide model response ➡️ Post-train to your sensor config + run batch inference ➡️ Validate every batch with autonomy-specific checks — obstacle correspondence and hallucination detection Full pipeline breakdown: appliedintuition.com/enginee…

Most AV testing runs on best-guess synthetic scenarios while 50 years of real crash data from @NHTSAgov's Fatality Analysis Reporting System (FARS) sits untapped. The data surfaces high-level patterns that engineers can use to fill in the gaps to reconstruct an incident, then multiply it into dozens of variations. Risk isn't evenly distributed. Our testing shouldn't be either. This is the approach we've built at Applied Intuition: treat the crash database not just as a record of the past, but as a blueprint for the future. We have the data and the technology. appliedintuition.com/blog/cr…

We're building physical AI for every moving machine. 🎧 Tune into the full @latentspacepod episode: piped.video/watch?v=rv23_KcH…

“Every automaker is investing in AI and software-defined vehicles, but many are still running on development processes built for a different era.” Live on stage at last week's AWS Summit Japan, engineers demoed Applied Intuition's AI-power Vehicle OS development environment on a Nissan Leaf demo vehicle, making a change to a vehicle welcome sequence and deploying it over-the-air in real time. What used to take months of automotive software development can now happen in minutes. appliedintuition.com/blog/ni…

"The overall ratio I look at is person to magnitude of impact." Our Deputy CTO @malharhar breaks down why fast-growing, mid-sized companies are the sweet spot for engineers looking to maximize their agency and scale. appliedintuition.com/blog/ma…

Autonomous haul trucks operate with no lane lines, no curbs, no signs - just berms, unmaintained haul roads, and obstacles of every size, shape, and material composition. A single sensor isn't enough. Here's how layered sensors help our trucks see in any condition: ➡️ Lidar: 360° coverage in low to no visibility - but some materials absorb rather than reflect, producing returns too weak to register as obstacles ➡️ Cameras: enrich the perception stack beyond what lidar alone can convey, and compensate for each other's blind spots in low light ➡️ Radar: measures the radial velocity of moving objects through the Doppler effect, giving the system the data it needs to assess whether a collision course is developing - something neither lidar nor cameras can do well Our SDS stack is built to a different standard. When the system encounters uncertainty, it doesn't default to a hard stop, instead it plans a path around the obstacle. Learn more about the sensor stack behind our self-driving system: appliedintuition.com/blog/au…

World foundation models can generate photorealistic, physically grounded sensor data for scenes that were never driven: any weather, any lighting, any ODD. Applied Intuition has built a complete toolchain around world foundation models that lets autonomy developers turn their fleet data into diverse, validated sensor datasets for production use. We've built a reference implementation around @nvidia Cosmos world foundation models and are excited to share this with developers. The pipeline runs in five stages, end-to-end: ➡️ Curate and auto-label fleet data into conditioning-ready segments ➡️ Extract scenario, map representations, and model sensor geometry to ground the model in the real world ➡️ Generate recipes, conditioning and prompts to guide model response ➡️ Post-train to match sensor configurations and run model inference in batch on cloud ➡️ Validate every batch with autonomy-specific checks and image quality metrics The model is powerful. The pipeline is what makes it usable in production. Read the full pipeline breakdown here: appliedintuition.com/enginee…

Autonomy doesn’t scale when every program builds its own infrastructure. During @SecondFront Offset Symposium, our Head of Federal Growth, Ezra Shapiro spoke about our work with the @CDAODoW on Autonomy Factory to build an enterprise autonomy pipeline that accelerates autonomy development and deployment across the @DeptofWar at mission speed. Learn more about how we’re accelerating the Autonomy Industrial Base: appliedintuition.com/autonom…

本日、Applied Intuitionの自動運転システムが正式に日本へ上陸しました。日本の自動車業界は、サプライヤーに対して独自の基準を課しています。 私たちは長年にわたり、日本に拠点を持つチーム、日本でのデータインフラ、日本の企業とのパートナーシップを構築し、その基準を満たす準備を進めてきました。 自動運転を世界中でスケーラブルにするためのこの重要な一歩について、詳しくはこちら：appliedintuition.com/ja/blog… Today, Applied Intuition's Self-Driving System officially comes to Japan. The Japanese automotive industry holds its suppliers to a different standard. We've been building to meet it for years: local teams, local data infrastructure, local partnerships. Learn more about this important step toward making autonomy scalable worldwide: appliedintuition.com/blog/ap…

It’s no longer about just banging out code. As our co-founder and CTO, Peter Ludwig, said on @latentspacepod, the engineering talent standing out the most today knows how to ask the right questions and integrate AI tools into their workflow effectively. Want to build with us? Learn more about our team: appliedintuition.com/careers… Watch into the full episode here: piped.video/watch?v=rv23_KcH…

What does it take to be Deputy CTO at Applied Intuition? For @malharhar: solving the hard problems. Watch the full interview here: appliedintuition.com/blog/ma…

Build physical AI that moves the world. Join our team. appliedintuition.com/careers…

エンド ツー エンドの自動運転ソフトウェアから次世代 HMI システムに至るまで、Applied Intuition は自動車業界における自動運転技術の普及を推進しています。 人とくるまのテクノロジー 2026 横浜にて、弊社ブースにお立ち寄りいただいた皆様、誠にありがとうございました。 From end-to-end autonomy software to next-generation HMI systems, we are scaling autonomy for automotive. Thank you to everyone who stopped by our booth at this year’s Automotive Engineering Exposition at the Yokohama Expo.

Can VLA models for autonomous driving perform without depending on complex reasoning or massive datasets? Our new approach, NoRD (No Reasoning Driving), proves you don’t need massive datasets or heavy reasoning for high performance. ➡️ 60% reduction in training data ➡️ 3x token efficiency ➡️ Competitive benchmark results By eliminating the reasoning overhead, NoRD creates a recipe for high-performance autonomous driving that is not only cheaper to train but also faster to develop and deploy. Read the research blog here: appliedintuition.com/researc… Want a deeper dive? Learn more here: nord-vla-ai.github.io/

Our team is heading to Denver for #CVPR2026 to share the work we are doing to build the future of physical AI. From research to build, come see how we are translating cutting-edge CV into real-world impact. 📍 Find our team of researchers and engineers at Booth #639. Read some of our recent research here: appliedintuition.com/researc…

"Applied Intuition's software architecture is designed to allow different brands to maintain distinct identities even while sharing the same underlying technology". @freep freep.com/story/money/cars/s…

Current world models often separate spatial reasoning from temporal modeling, leading to consistency bottlenecks and scaling limits. We developed RAYNOVA, a 4D world foundation model, to address this by unifying space and time into a single representation via a pure auto-regressive framework. The result: Data-driven high-fidelity simulation data that scales across any vehicles, anywhere. Read the full technical breakdown: appliedintuition.com/researc…