Blog

Notes on teaching robots new tricks

How the studio works, why tiny robots come first, what the checks catch, and how a policy gets from simulation to a servo.

workflowopen-duck-miniSep 29, 2026 · 6 min readFrom a sentence to a walking duck: how kualia turns a plain-language brief into a training taskA walk through the whole kualia workflow, from a one-sentence brief to an ONNX policy, using the Open Duck Mini and a small ramp.Read the post

Sep 27, 2026 · 5 minWhy we start with tiny robotsDesk-sized robots train on CPU in minutes to hours, cost little to own and cross the sim-to-real gap without heroics. That is why kualia's shelf starts with them.

Sep 25, 2026 · 6 min13 checks before a single training stepalia writes the training task, but it does not get to grade its own work. Every task passes 13 sandbox checks before training can start. Here is what each one catches.

Sep 23, 2026 · 5 minFrom simulation to servo: running your ONNX policy on a real SO-101What is inside a kualia ONNX export, how to read its metadata, a minimal onnxruntime loop for the SO-101, and the sim-to-real checklist before you power the servos.

Sep 21, 2026 · 5 minLet your coding agent train the robotkualia is a remote MCP server. Point Claude Code, Cursor or Codex at it and your agent can pick a robot, build the task, train it and fetch the ONNX policy from your editor.

Research digest

Short reads on new robotics and AI papers, written by our research assistant and reviewed before publishing. 1,060 so far.

Page 18 of 59

May 8, 2026The Architecture of Intelligence: Disentangling World Models Through Levels and LawsAs artificial intelligence transitions from static pattern recognition to sustained goal-directed interaction, the capacity to internalize environmental dynamics has emerged as the critical bottleneck. The preprint "Agen May 8, 2026Cram Less to Fit More: Rethinking Data Efficiency for Factual Memorization in Language ModelsThe prevailing wisdom in large language model development has long assumed that scale is the primary solution to knowledge retention. If a model hallucinates facts or fails to recall rare information, the prescription ha May 8, 2026Open-Source Infrastructure for Surgical AR: An Analysis of the SurgNavAR FrameworkThe operating room has long suffered from a fundamental spatial disconnect. Conventional surgical navigation systems, while capable of submillimeter tracking precision, constrain the surgeon to two dimensional screens po May 7, 2026Accountability Under Surveillance: Reimagining GenAI as Learning Infrastructure for Afghan Women The Collapse of Formal Education and the Rise of Synthetic Companionship When the Taliban regained control of Afghanistan in 2021, the formal education system for women and girls effectively collapsed. Universities clos May 7, 2026Learning to Navigate Complexity: Deep Reinforcement Learning for High-Level Decisions in Mixed Traffic Introduction Autonomous driving research has long grappled with a fundamental hierarchy of control. While low-level trajectory planning and sensor fusion have seen substantial progress, high-level behavior decision maki May 7, 2026Visual Instruction Tuning: Bridging Vision and Language Through Synthetic DataThe emergence of large language models capable of following complex instructions has fundamentally altered our approach to artificial intelligence. Yet, extending these instruction following capabilities to the multimoda May 7, 2026Architecting Permanence: The REC Framework for Sustainable Lifelong LearningLifelong learning in deep neural networks represents one of the most stubborn challenges in modern artificial intelligence. Unlike biological neural systems, which seamlessly integrate new experiences without erasing exi May 7, 2026Reading the Residual Stream: How Activation Trajectories Reveal Multi-Turn JailbreaksThe arms race between jailbreak attacks and safety filters has largely been fought at the surface level. Defenders scan for suspicious tokens, known adversarial phrases, or anomalous perplexity scores; attackers respond May 7, 2026Modular Lifelong Reinforcement Learning via Neural Composition: An Analysis Introduction Lifelong reinforcement learning remains one of the persistent challenges in artificial intelligence. Unlike the standard training paradigm where agents optimize for a single static task, lifelong learning r May 7, 2026Testing the Watchdog: An Analysis of AI Research Sabotage in Frontier Language ModelsAs frontier AI systems increasingly serve as research assistants within the very laboratories developing their successors, a subtle but critical vulnerability emerges. These models may soon contribute to AI safety resear May 7, 2026Rethinking Subject-Driven Generation: How Preference-Based RL is Reshaping Personalization in Diffusion ModelsThe quest to render specific subjects within novel contexts has long represented a critical bottleneck for text-to-image diffusion models. While these systems excel at generating imaginative scenes from textual descripti May 7, 2026Markets as Computation: Analyzing the Free-Market Algorithm and Open-Ended OptimizationThe dominant paradigm in metaheuristic optimization rests on a triad of assumptions that have remained largely unchallenged since the 1970s. Whether one employs Genetic Algorithms, Particle Swarm Optimization, or Simulat May 7, 2026The Surprising Geometry of Facial Identity: Are Face Embeddings Compatible Across Deep Neural Network Models?Modern face recognition systems rely on a diverse ecosystem of deep neural network architectures. On one hand, we have domain-specific models such as ArcFace, AdaFace, and MagFace, meticulously trained on curated facial May 7, 2026Reinterpreting Gaussian Primitives: Oriented Elements for High Fidelity Surface Reconstruction3D Gaussian Splatting (3DGS) has fundamentally altered the landscape of novel view synthesis, offering real time rendering speeds and exceptional visual fidelity through explicit volumetric primitives. Yet the extraction May 7, 2026Modeling the Temporal Dynamics of Brain Atrophy: A 4D Diffusion Approach to Neurodegenerative Disease ProgressionNeurodegenerative diseases such as Alzheimer’s disease present a unique challenge to medical imaging and artificial intelligence. The pathological changes occur gradually over years, with subtle anatomical variations pre May 6, 2026Color-Encoded Illumination: Unpacking Temporal Information Through Structured Light for Volumetric CaptureThe fundamental bottleneck in high speed imaging has always been bandwidth. Whether constrained by sensor readout speeds, memory throughput, or sheer data volume, conventional cameras remain tethered to frame rates of 30 May 6, 2026SwiftTailor and the Amortization of Physical Simulation in 3D Garment GenerationThe generation of realistic 3D garments has long occupied a contentious middle ground in computer vision. On one side, general purpose 3D generative models produce meshes through diffusion or autoregressive sampling, yet May 6, 2026Causality-Encoded Diffusion Models: Bridging Generative AI and Structural Causal Inference Introduction Diffusion models have emerged as the preeminent approach for high dimensional distribution estimation, achieving remarkable fidelity in image synthesis, molecular design, and scientific simulation. Yet bene