AI foundations · Agent architecture
Model, harness, agent: what are we actually talking about?
A dinner-table debate about Cursor reveals why these terms describe different layers of the same system.
Understand the three layersGuides and insights from AIConnector.ai
Understand the tools. Make informed engineering choices.
We explore AI tools, agent architecture, evaluation, and knowledge systems to help developers and engineering teams understand their options.
Featured
Field guides
Evidence-led comparisons and explainers for decisions that look simple on the surface but have materially different architecture, cost, and control implications.
Agent architecture · Memory systems
Agent memory: what survives the next session?
Understand memory and dreaming, then follow a lesson through verification, storage, retrieval, and evaluation.
Read the memory guideNew field guide · 6 November 2026App idea.
Working prototype.
Working prototype.
AlloyDB Express · Application prototyping
AlloyDB Express: Postgres in ~2 seconds for developers and agents
Building prototypes? Explore fast Postgres provisioning and scale-to-zero compute, with a practical walkthrough for developers and agents.
Read the deep dive
New field guide · 5 November 2026Analytics.
Inside chat.
Inside chat.
MCP Apps · Analytics architecture
MCP Apps: bring analytics into the conversation
Deliver focused analytics widgets in an agent’s chat. Understand the benefits, architecture, and when a full dashboard still makes sense.
Read the field guide
New field guide · 3 November 2026Your agent.
Your architecture.
Your architecture.
Agent architecture · Implementation guide
Beyond Dots and Duet: build your own portable agent
Own task state, memory, tools, and approvals. Explore a durable architecture and the tests that prove you can change model providers.
Read the architecture guideEngineering landscape · 3 November 2026Rethinking Git.
Source control · AI infrastructure
Why companies are rethinking Git
Cursor’s Origin and Continuity, Scion, GitLab, and OpenAI’s reported project: compare confirmed systems, developing projects, and historical precedents.
Read the source-control guide
New field guide · 2 November 2026When to delegate.
Agent architecture · Decision guide
When does your agent actually need subagents?
Choose a shared loop, skills, or delegated work using concrete examples, a decision table, and a repeatable evaluation plan.
Read the decision guide
Agent architecture · Context engineering
How a turn works in an agent harness
Why model calls are stateless, why context and token usage grow, how tools create request loops, and what compaction changes.
Read the explainer
Project Atlas
GraphRAG
Knowledge systems · Architecture landscape
GraphRAG feature and pricing matrix
A practical comparison of eight open-source graph-knowledge approaches, production reference architectures, deployment trade-offs, and the real cost model behind them.
Explore Project Atlas
Agent continuation
Different clocks.
Agentic engineering · Control models
Autopilot, Loop, and Goal
A vendor-neutral field guide to three continuation mechanisms—what makes each agent continue, what makes it stop, and how to choose by the required stopping rule.
Open the field guideArchive
Earlier explorations
AI research · Plain-language guide
Jev, explained.
AI research · Model evaluation
What are Jev, AUROC, and RLCD?
A beginner-friendly guide to how Jev checks AI responses, how AUROC measures its ranking, and what is publicly known about RLCD.
Read the explainer
Applied AI · Evaluation practice
Build the loop.
AI agents · Evaluation and quality
Build an evaluation loop for your AI agent
A practical guide to test cases, grading rubrics, skill improvement, bias checks, and enterprise monitoring—with a customer support example.
Read the field guide
Applied AI · Enterprise operations
AI-powered process optimization
An exploration of how intelligent systems can identify bottlenecks, improve decisions, and make operational workflows more adaptive.
Read article
Machine learning · Exploration
Understanding torque clustering in AI
A technical introduction to a physics-inspired approach to identifying structure in complex, multidimensional data.
Read article