AI Concepts

A protocol referenced as needing redesign for agent-first usage. In this newsletter it is grouped with APIs and CLIs as software interfaces that must become more discoverable and forgiving for AI agents.

MCP is emerging as a key protocol for connecting AI agents to tools, systems, and external context.

Autonomous or semi-autonomous software systems that can act across tools and workflows. The newsletter frames agents as buyers, tool consumers, and the primary audience for protocols like MCP.

AI agents are increasingly framed as autonomous operators that can use tools, files, APIs, and workflows on behalf of users.

An AI development pattern where models act more like autonomous coding agents. The newsletter uses it to describe both NVIDIA Dynamo’s target workload and GPT-5.5/Codex improvements.

Agentic coding describes AI systems that act more like autonomous software agents than passive coding assistants.

A rapid, intuition-driven way of building software with AI assistance. For PMs, it represents low-friction prototyping and UI iteration.

Vibe-coding is an intuition-driven, AI-assisted way to build software through prompts, iteration, and lightweight debugging.

A method for structuring prompts and surrounding artifacts across multiple layers, such as specs, wireframes, and data, to improve AI output quality. It is especially useful for PMs designing AI-assisted product workflows.

Context engineering extends beyond prompt writing to include specs, memory, retrieval, tools, and structured data.

A pattern for answering questions by retrieving relevant context and generating responses from it. The newsletter highlights multimodal RAG for searching across audio, image, and video data.

RAG grounds model outputs in retrieved external context, making it a foundational pattern for enterprise AI assistants and search products.

Agents that perform coding tasks and can increasingly orchestrate adjacent workflows like design. The newsletter uses them as the execution layer for Design.md scripts.

Coding agents are evolving from code generators into autonomous workflow executors that can analyze, write, debug, and document software.

A concept for modular agent capabilities or instructions, mentioned as an emerging hint toward open standards. It is discussed alongside agents.md in the context of agent harness interoperability.

Skills describes modular, reusable agent capabilities or instructions that may work across tools and environments.

Simon Willison’s command-line LLM tool for interacting with models and APIs. This release adds support for OpenAI’s Responses endpoint and better reasoning-token handling.

LLM here refers both to large language models broadly and specifically to Simon Willison’s `llm` command-line tool in the newsletter context.

A technique for grounding model outputs in retrieved information. It is cited here as a component of a modular agent framework.

RAG improves model outputs by retrieving relevant external information at generation time.

A workflow/mode for using AI systems to search the web, synthesize information, and produce detailed reports. The newsletter frames it as a practical capability for research-heavy PM work.

Deep Research combines web search, multimodal input handling, synthesis, and report generation into one AI workflow.

A workflow pattern where a main AI system delegates parts of a task to parallel helper agents. Relevant to PMs because it can improve speed, context management, and long-running task execution.

Subagents let a main AI system delegate parts of a task to specialized helper agents, often in parallel.

Benchmarking methods for evaluating AI coding agents in realistic software tasks. The newsletter notes that infrastructure variability can materially affect scores.

Agentic coding evals measure AI coding agents on realistic software tasks rather than isolated prompt-response coding tests.

An open-source agent framework associated with Harrison Chase. In the newsletter it is being optimized for open-source models as closed-model costs rise.

Deepagents emerged as an open-source Claude Agent SDK tied to Harrison Chase and the LangChain ecosystem.

The class of models discussed as having a blind spot with continuous, high-dimensional, noisy data. This concept is used to frame a limitation in current AI capabilities.

LLMs are foundational to many generative AI products but are best understood as powerful language models, not universal intelligence systems.

The practice of building software systems where agents plan and execute tasks with autonomy. The newsletter uses it in the context of anti-patterns and agent behavior management.

Agentic engineering centers on building software with agents that can plan, write, and execute code autonomously.

A paradigm that treats cloud infrastructure as autonomous coding agents to automate deployment and operations. For AI PMs, it reframes infrastructure as an agentic workflow rather than a static system.

Agentic Infrastructure treats cloud operations as workflows executed by autonomous coding agents rather than static systems.

Reusable Claude-based skill modules that package agentic workflows into portable components. The newsletter frames them as a way to avoid building AI agents from scratch.

Claude skills package repeatable AI workflows into reusable modules, reducing the need to build agents from scratch.

A security risk pattern where AI agents have private data access, ingest untrusted content, and can exfiltrate data. For AI PMs, it is a key framework for designing safe agent features.

The lethal trifecta describes the dangerous combination of private data access, untrusted content ingestion, and exfiltration capability in one AI system.

A practice for testing whether AI agents complete tasks reliably across real inputs, edge cases, and versions using output, trajectory, and component-level checks.

Agent evaluation measures whether AI agents complete tasks reliably across real inputs, edge cases, and version changes.

A test-driven development pattern adapted for coding agents. It emphasizes an iterative failure/success loop that can make agentic coding more reliable.

red/green TDD adapts classic test-driven development into a workflow that makes coding agents validate their output step by step.

A lexical retrieval ranking function used here to select relevant tool definitions. In PM tooling, it helps improve retrieval accuracy and reduce context-window bloat.

BM25 is a lexical ranking function that helps retrieve the most relevant documents or tool definitions for a given query.

A collection of techniques and patterns for building agentic systems. The newsletter frames it as a guide page for AI builders.

Agentic Engineering Patterns is a guide that collects practical techniques for building and operating agentic systems.

An approach to AI systems where agents perform tasks autonomously with tools and browser interaction. The newsletter frames 2026 as a year focused less on novelty and more on trust in deployed agentic systems.

Agentic AI describes systems that can autonomously complete multi-step tasks using tools, APIs, and browser interaction.

A framework for defining, managing, and retiring capabilities that AI agents can use. The newsletter frames it as an operational way to keep agent behavior current and useful.

Agent Skills break agent behavior into manageable, reusable capabilities instead of relying on one monolithic prompt.

A software architecture paradigm where engineers orchestrate agents instead of hard-coding decision trees. For PMs, it suggests product teams may design systems around LLM behavior rather than deterministic logic.

Agent-first software design shifts software building from hard-coded decision trees to orchestrated agent behavior.

A legacy programming language often targeted for modernization and migration efforts. For PMs, it represents enterprise technical debt and transformation risk.

COBOL remains core to many enterprise systems despite being viewed as legacy technology.

Systems composed of multiple cooperating AI agents, often designed to divide work and collaborate through structured patterns. The newsletter references building these systems with Python and agent-to-agent communication patterns.

Multi-agent systems divide work across specialized AI agents that coordinate through structured communication patterns.

OpenTelemetry is an observability standard for traces, logs, and metrics. The newsletter mentions Codex exporting agent-aware telemetry through it for auditing and monitoring.

OpenTelemetry is a standard for collecting traces, logs, and metrics across software and AI systems.

A programming language commonly used for building AI systems and agent workflows. The newsletter references it in the context of constructing multi-agent systems from scratch.

Python is the dominant implementation layer for modern AI experimentation, orchestration, and agent workflows.

A tool interface used with skill.md to reduce token usage and run MCP commands in a more efficient way.

CRI is a lightweight interface for running MCP commands with lower token overhead.

AI models whose weights or availability are open enough to encourage broad reuse and experimentation. The newsletter frames them as a driver of innovation across the ecosystem.

Open models are framed in the newsletter as a major driver of AI innovation across startups, researchers, students, and industries.

A memory architecture pattern for AI agents that separates different memory layers to improve context retention and task performance. It is presented as part of the design of autonomous coding assistants.

Layered memory separates short-term, task-level, and longer-term memory to improve AI agent performance.

The process of updating legacy COBOL systems, often for enterprise migration and maintenance. AI agents are increasingly positioned as tools to accelerate this high-friction modernization work.

COBOL modernization focuses on updating legacy mission-critical systems for maintainability, integration, and migration.

Leading AI labs that control high-demand model APIs and compute. The newsletter uses the term to describe vendors that might restrict API access to prioritize their own products and customers.

Frontier AI labs control highly demanded model APIs and may prioritize their own products or top customers when compute is scarce.

A structured-prompt framework for improving the consistency and quality of outputs from Claude Code. It is positioned as a way to turn an AI coding assistant into a more reliable development partner.

SuperClaude is a community framework that uses structured prompts to improve the consistency of Claude Code outputs.

A defensive technique mentioned as part of Claude Code's strategy to deter model distillation by misleading competitors' training runs.

Anti-distillation poison pills are designed to make model outputs less useful for competitors attempting distillation.

A PM framework focused on user value, tradeoffs, and outcomes rather than just technical implementation. Mentioned here as a skill engineers should develop in AI product teams.

Product-thinking emphasizes user value, tradeoffs, and outcomes over pure implementation.

A protocol for connecting AI models to external tools and servers. The newsletter references discovery of MCP servers and reducing MCP token usage.

Model Context Protocol standardizes how AI models and agents connect to external tools, servers, and data sources.

A lightweight skills-based pattern for packaging agent capabilities in small context-efficient files.

skill.md is a lightweight pattern for packaging agent capabilities into small, context-efficient files.

An architecture where multiple specialized agents collaborate instead of one general-purpose agent. The newsletter includes debate over whether this is necessary versus using a single tool-loaded agent.

A multi-agent system uses several specialized agents to collaborate instead of relying on one general-purpose agent.

A practice of capturing learnings from prompts and agent interactions to steadily improve system behavior over time. For PMs, it is a feedback-loop mindset for iterative AI product improvement.

Compound Engineering is the practice of capturing prompt and agent learnings so future AI outputs improve over time.

A pattern for agent-to-agent communication and collaboration. The newsletter mentions it as part of a step-by-step approach to building multi-agent systems.

A2A is a pattern and protocol framing for how multiple AI agents communicate and collaborate.

A training approach used here to teach Composer to self-summarize, reducing reliance on handcrafted prompts.

Reinforcement learning helps AI systems improve behavior based on outcome feedback rather than prompts alone.

Programmable interfaces that let AI agents and software systems access services and complete tasks. The newsletter positions APIs as one of the means for agents to act on behalf of users.

APIs let AI agents access services and take actions on behalf of users.

The practice of connecting agents to external developer tools such as linters and debuggers. It is highlighted here as a building block for effective coding agents.

Tool integration connects AI agents to external developer tools such as linters, debuggers, and test runners.

A search tool mentioned as part of ingesting PM work into Claude Code. It appears to support retrieval over a large personal knowledge base.

QMD appears to be a search and retrieval layer used to access large personal or operational knowledge bases.

A modular layer that adds tools, guardrails, and custom instructions to AI agents. It is described as a composable harness for production agent systems.

Agent middleware is a modular layer for adding tools, guardrails, and instructions to AI agents.

An agent design pattern where work is split into sub-tasks and assigned dynamically. In the newsletter, it is one of the core ingredients for building autonomous coding agents.

Task delegation breaks complex agent objectives into smaller sub-tasks assigned dynamically across tools or specialized components.

Attack technique where malicious prompts manipulate AI systems or agents. Here it is connected to a GitHub issue triage workflow exploit.

Prompt injection manipulates AI systems by embedding malicious instructions in untrusted inputs the model consumes.

A product and engineering concept describing the hidden cost of AI-accelerated development when teams lose shared understanding of the system. It reframes debt from code maintenance to team cognition and system comprehension.

Cognitive debt describes how AI-accelerated development can shift costs from code maintenance into lost team understanding.

A test introduced by Andrew Ng for evaluating economic utility. It is framed as a way to assess whether AI systems provide meaningful real-world value.

The Turing-AGI Test evaluates AI progress based on economic utility rather than abstract intelligence claims.

A framework for specifying goals, context, and guardrails in multi-agent systems. It helps PMs guide autonomous agents with explicit objectives and stop rules rather than rigid control.

Intent Engineering helps PMs specify objectives, context, and guardrails for autonomous agents.

A concept covering how organizations evaluate large language models consistently and meaningfully. The newsletter frames standardization of benchmarks as a major enterprise challenge.

LLM benchmarks give organizations a repeatable way to compare model performance on real product tasks.

AGI is referenced as the frontier toward which current AI development is moving. In PM terms, it frames long-term product strategy, governance, and risk discussions.

AGI is a strategic concept that shapes AI roadmaps, governance, and stakeholder expectations more than a single agreed technical milestone.

A workflow framework for building customizable agentic systems. It is highlighted as integrating with ACP.

Agent Workflows is a framework for building customizable agentic systems within the LlamaIndex ecosystem.

A memory architecture that mimics human memory instead of relying on RAG or vector search. For PMs, it suggests alternative approaches to long-context recall and personalization.

Large Memory Models are described as a memory architecture that mimics human memory rather than relying on RAG or vector search.