Blog
Notes from building api0 and our Rust tools: AI agents, MCP, Azure integration and Rust itself.
Filter by topic
Showing 6 of 65 articles • Page 1 of 11
AI in Companies Isn't 50 Use Cases. It's 3 Things.
Most generative and agentic AI projects fit into three categories: agents for customers, integration layers for systems, and copilots for employees. Here is why that lens beats a list of fifty use cases.
Your Logic Apps Are a Graph, and the Azure Portal Only Shows You a List
Past a dozen workflows, an Azure Logic Apps estate stops being a list and becomes a graph connected by Service Bus queues and Event Grid topics. Here is why the portal cannot show you that graph, and what to do about it.
How Far Did That Order Get? Tracing a Correlation ID Across Cosmos DB
A customer says their order never arrived. You have a correlation id and an Azure Cosmos DB account with more containers than anyone remembers. Here is why the empty results matter more than the full ones.
GC Pauses and Latency: The Hidden Cost of High-Level Languages
Java, Python, and JavaScript offer convenience, but garbage collection introduces unpredictable latency. Explore how runtime memory management affects performance in real systems.
C Gives You Control, But at What Cost?
C avoids garbage collection and gives manual memory control, but opens the door to dangerous bugs. Explore real-world memory issues and why they matter.
Rust: Memory Safety Without Garbage Collection
Rust gives you the performance of C with memory safety enforced at compile time. Learn how ownership and borrowing eliminate entire bug classes.
Episode 0: Why Code LLMs From Scratch
Understanding how LLMs work without a PhD in mathematics: what this series promises, what you will learn, and the one idea that ties every episode together.
Episode 1: Attention Is Just 4 Mathematical Operations
Behind Transformers and the attention mechanism there are only four simple operations: matrix multiplication, transposition, scaling and softmax. Here they are, in Rust.
Episode 2: Q, K, V — Why 3 Matrices Instead of One?
Query, Key, Value: why real models project each word into three different spaces, and what that separation of roles changes for attention.
Episode 3: Embeddings — Absolute at the Start, Contextual at the End
Is an embedding the word on its own or the word in context? Both, at different moments. Plus the three engineering details around attention: multi-head, masking and position encoding.
Episode 4: Transfer Learning — Reusing Existing Knowledge
Why nobody trains a model from zero: where transfer learning comes from, what it saves you, and its three strategies — feature extraction, fine-tuning and LoRA.
Episode 5: The Key Concepts Around Transfer Learning
Pre-training, fine-tuning, catastrophic forgetting, freezing, domain shift and learning rate: the transfer learning terms you hear all the time, explained simply with an analogy for each.