Machine Learning Foundations: An Interactive Guide
A new interactive guide that builds modern AI up from a straight line to a transformer: gradient descent, backpropagation, embeddings, and attention, with every chart explorable.
Most Read
Requirements as Code: Git-Native Business Documents for Agentic Workflows
Exploring the idea of putting business requirements, architecture diagrams, and domain models in Git — and how this could enable agentic pipelines from requirement change to deployed code.
February 26, 2026Benchmark
LLM Trading Decisions →
Which language models make the best crypto-futures trading calls, and under what market conditions? A deterministic backtest of agentic vs. single-call LLM traders across 20 historical scenarios, measured against buy-and-hold, technical-analysis, and random baselines.
View the benchmark →What's Happening
Analysis
Production Traces Are Becoming the Training Signal for Agent Fixes
A cluster of releases in late June 2026 shows agent improvement loops moving from offline eval suites to continuous systems that mine production traces, cluster failures, and propose code changes.
July 4, 2026The Harness Stops Being Static: Self-Modification, Trace-Mined Updates, and Per-Model Profiles
Multiple converging releases suggest the agent harness is becoming a runtime-mutable artifact that agents, traces, and model-specific tuning all reshape — with concrete consequences for how teams version and govern infrastructure.
June 27, 2026Budget Becomes a First-Class Runtime Concern
Spend caps, time budgets, and cost-aware training are converging into a distinct governance layer between agents and the models they call.
June 20, 2026Invisible Model-Side Guardrails: What Silent Routing Means for Agent Reliability
How model-side safety classifiers and silent fallback routing affect evaluation, tracing, and cost accounting in production agent systems.
June 13, 2026Per-Model Harness Profiles and Runtime-Authored Workflows
Two converging shifts — per-model harness profiles and agents that write their own orchestration at runtime — are breaking the assumption that an agent harness is a stable, model-agnostic abstraction.
June 6, 2026The Agent Becomes the Optimization Unit
Multi-agent systems are being instrumented and tuned at the agent level: credit assignment, failure-mode decomposition, and policy learning all treat the individual agent as the thing you measure, replace, or evolve.
May 30, 2026From the Archive
Scaffold Choice Dominates Model Choice in LLM Trading Backtests
Across 540 crypto futures backtests, switching from a single-prompt agent to a ReAct-style loop produced a larger performance swing than switching models — and most configurations still lost to a passive index baseline.
May 27, 2026The Sandbox Becomes a Runtime Primitive
Isolated code execution environments are emerging as a distinct layer of the agent stack, separable from both the harness and the model — with implications for security, portability, and cost.
May 23, 2026Context as a Deployable Artifact: The Third Layer of the Agent Stack
Agent context files are being pulled out of repos and into versioned, governed runtime stores — creating a third deployment surface alongside harness code and model weights.
May 16, 2026