Platform · Collections

Curated Collections

Carefully assembled reading paths that connect articles, projects, and labs around coherent technical themes — structured for depth, not just discovery.

Collections

Items

Featured

AI Agent Orchestration

How to design, coordinate, and scale multi-agent systems — from context management to tool routing and memory architectures.

3 articles1 projects1 labs5 total

Why coordination, not raw inference, is the real bottleneck in agentic systems.
Short-term, episodic, and semantic memory — trade-offs and implementation patterns.
MCP, A2A, and the emerging interoperability layer between AI systems.
+ 2 more

Explore collection

Autonomous Trading Systems

Building systematic, AI-driven trading engines — signal generation, risk modeling, execution architecture, and evaluation frameworks.

2 articles1 projects1 labs4 total

Memory patterns that apply directly to stateful trading agents.
Retrieval strategies for market context — when graphs beat embeddings.
The full trading engine — architecture, signal model, and execution layer.
+ 1 more

Explore collection

Modern DevOps Stack

Infrastructure, deployment automation, and operational practices for engineering teams shipping at scale. From CI/CD architecture to observability and incident response.

3 articles2 projects1 labs6 total

Observability-first infrastructure design
Opinionated toolchain evaluation for 2026
Performance budgets as engineering discipline
+ 3 more

Explore collection

All Collections

Production TypeScript

Patterns for writing TypeScript that actually holds up in production — type-safe APIs, resilient state management, and React 19 architecture.

3 articles3 total

Layered API design with typed error boundaries and retry policies.
Practical concurrent React — useTransition, Suspense, and deferred updates.
End-to-end TypeScript MCP server: tools, resources, and prompts.

Explore collection

Generative AI Tools

Building production-ready generative AI interfaces — from 3D geometry synthesis to structured output pipelines and prompt engineering.

1 articles1 projects1 labs3 total

MCP tool servers: the integration layer for generative AI.
Text-to-3D: LLM intent parsing → parametric synthesis → printable mesh.
Live demo — describe a shape, preview and export in the browser.

Explore collection