The First AI-vs-AI Breach: How an Autonomous Agent Attacked Hugging Face and Why Open-Weight Models Saved the Day
Something we’ve all been predicting finally happened. Hugging Face disclosed last week that they were hit by a security intrusion
From Our Blog
The Strangler Fig Pattern: Incrementally Migrating Legacy Systems Without the Big Rewrite
Every engineering team eventually faces the same daunting question: what do we do with the legacy monolith? The temptation is
Event Sourcing in Practice: Building a System That Never Forgets
Most systems today use CRUD: read the current state, modify a row, write it back. The database remembers where things
Continue readingEvent Sourcing in Practice: Building a System That Never Forgets
Xiaomi-Robotics-1: When Scaling Laws Finally Arrive in Robotics
Robotics has a data problem. While language and vision models have ridden scaling laws to ever-higher capabilities, robot learning has
Continue readingXiaomi-Robotics-1: When Scaling Laws Finally Arrive in Robotics
Kimi K3: Moonshot AI’s 2.8 Trillion Parameter Open-Source Behemoth
Moonshot AI has just dropped Kimi K3, and it’s a monster. At 2.8 trillion parameters, it’s the world’s first open-source
Continue readingKimi K3: Moonshot AI’s 2.8 Trillion Parameter Open-Source Behemoth
Faster Docker Builds with BuildKit: Cache Mounts, Remote Caching, and Bake
Docker image builds are one of those things that quietly accumulate technical debt. You start with a simple Dockerfile, and
Continue readingFaster Docker Builds with BuildKit: Cache Mounts, Remote Caching, and Bake
Inkling: Thinking Machines Lab’s 975B Open-Weights Multimodal Model
The open-weights LLM landscape just gained a significant new entrant. Inkling, released on July 15 by Thinking Machines Lab, is
Continue readingInkling: Thinking Machines Lab’s 975B Open-Weights Multimodal Model
The Saga Pattern: Coordinating Distributed Transactions in Microservices
Distributed transactions are one of those problems that look simple until you split your monolith into services. In a single
Continue readingThe Saga Pattern: Coordinating Distributed Transactions in Microservices
Test Doubles Demystified: Stubs, Spies, Mocks, and Fakes in Go
Automated tests that hit real databases, live APIs, and production file systems are slow, flaky, and expensive to maintain. The
Continue readingTest Doubles Demystified: Stubs, Spies, Mocks, and Fakes in Go
Training-Inference Mismatch: Why Your LLM Reinforcement Learning Is Optimizing the Wrong Policy
Reinforcement learning has become the defining ingredient of modern LLM post-training. GRPO, PPO, and their variants drive the reasoning capabilities