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 59 of 59

Mar 13, 2026 Mar 13, 2026 · TR Mar 13, 2026Surface Light Field Tokenization: Learning 3D Appearance Through Strategic Subsampling Introduction The synthesis of realistic 3D objects requires more than accurate geometry. Real materials interact with light in complex ways, producing view-dependent effects such as specular highlights, Fresnel reflecti Mar 13, 2026 · TR Mar 13, 2026 Mar 12, 2026 · TRChamfer Distance'un Yapısal Çöküşü: 3B Nokta Bulutu Optimizasyonunda Gizli Çekici Nokta Giriş Üç boyutlu derin öğrenme uygulamalarında Chamfer distance, nokta bulutu (point cloud) rekonstrüksiyonu, şekil tamamlama (shape completion) ve 3B jenerasyon modellerinin vazgeçilmez kayıp fonksiyonları arasında yer Mar 12, 2026The Structural Failure of Chamfer Distance: Why Local Optimization Leads to Global Collapse Introduction Chamfer distance has become the de facto standard for training and evaluating point cloud reconstruction, completion, and generation systems. Its computational efficiency and geometric intuition, measuring Mar 12, 2026 · TR Mar 12, 2026Beyond Forward Vision: Enabling True Omnidirectional Awareness for Autonomous DronesMost autonomous drones today suffer from a fundamental sensory limitation. They navigate using forward facing cameras or limited field of view sensors, creating a dangerous blind spot that becomes critical during lateral Mar 12, 2026 · TRLLM'lerin Ürettiği C/C++ Kodunun Güvenlik Açısından İncelenmesi: LLM-CSEC Çalışması ve Geleceğe Dair Çıkarımlar Giriş Büyük dil modellerinin (Large Language Models, LLM) kod üretim kapasitesi, yazılım geliştirme pratiğini kökten değiştirme potansiyeline sahip. GitHub Copilot, ChatGPT ve benzeri araçlar artık günlük geliştirme akı Mar 12, 2026The Insecurity of Autoregressive Code Generation: Why Post-Hoc Analysis Fails and What Comes Next The Security Debt of Automated Programming The integration of large language models into software development workflows has fundamentally altered the economics of code production. Systems that once required hours of car Mar 12, 2026 · TRSurgFormer: Cerrahi Simülasyonda Ölçeklenebilir Deformasyon Öğrenimi ve Dinamik Bilgi Akışı Giriş Yüksek doğruluklu biyomekanik simülasyon, minimal invaziv cerrahi eğitiminden preoperatif planlamaya kadar geniş bir yelpazede kritik öneme sahiptir. Ancak geleneksel sonlu elemanlar yöntemi (Finite Element Method Mar 12, 2026Adaptive Information Routing in Learned Biomechanical Simulation: A Deep Dive into SurgFormer The Computational Bottleneck of Interactive Surgery High fidelity biomechanical simulation remains one of the most computationally demanding tasks in medical computing. Finite element methods (FEM), while physically acc Mar 12, 2026 · TRVision Transformers'ta Figür-Zemin Organizasyonunun Lokasyonu ve Editlenmesi: BEiT Modellerinde Gestalt Prensiplerinin Mekanistik Analizi Giriş: Global Receptive Field ve Algısal Örgütlenme Arasındaki Köprü Vision Transformers (ViT'ler), bilgisayarlı görü alanında evrimsel bir dönüşümü temsil ediyor. Dosovitskiy ve arkadaşlarının 2021 yılında ortaya koydu Mar 12, 2026Locating the Convexity Prior: Mechanistic Interpretability Meets Gestalt Psychology in Vision Transformers Introduction The internal computations of Vision Transformers remain largely opaque, despite their dominance in computer vision. While we can measure their performance on benchmarks, understanding how these models organ Mar 12, 2026Scaffolding Agency: Logical Pretraining for Aligned Reinforcement LearningThe persistent challenge of reward hacking in deep reinforcement learning reveals a fundamental tension in artificial intelligence research. When agents optimize for proxy objectives rather than intended goals, they exhi