Inside Vercel's Eve: A Deep Technical Walkthrough, with a Research Agent
What Eve is, every capability that matters, a full TypeScript research agent, and how it compares to LangChain's Deep Agents
Tutorials, deep dives, and project walkthroughs on building AI agents, RAG systems, and the full-stack infrastructure that ships them, with LangGraph, LangChain, and modern engineering.
Deep technical explorations of AI engineering.
What Eve is, every capability that matters, a full TypeScript research agent, and how it compares to LangChain's Deep Agents
A clear split between an agent's working memory and its cross-session memory, with LangGraph code for checkpointers, stores, and the semantic/episodic/procedural breakdown.
Deep Agents bakes planning, context management, subagents, and human approval into the agent loop. Here is why it exists, every main feature, and a production-grade customer support agent built with it.
Open-source tools and apps I've built. All open source.
A tiny async framework for reliable, observable tool-using agents in Python, small enough to read in one sitting.
Python 3.11+ · asyncio · OpenAI SDK · PyYAML · uv
Search the entire UK Register of Licensed Sponsors: 125,000+ verified companies, updated from GOV.UK, in one fast, filterable interface.
Next.js 16 · React 19 · TypeScript · MongoDB Atlas · Mongoose · Tailwind CSS v4 · shadcn/ui · Framer Motion · Railway
An AI study companion that blends smart scheduling, task management, and personalised tutoring: five AI agents, each with its own teaching style.
React 18 · TypeScript · Vite · React Router v7 · Tailwind CSS v4 · Zustand · Motion · Python · Railway
From first principles to production systems.
Building, coordinating, and shipping AI agents.
BrowseAI InfrastructureThe infrastructure under AI systems: vector stores, serving, scaling, and cost.
BrowseLLM IntegrationWiring LLMs into real software: APIs, structured outputs, tools, and reliability.
BrowseRAGRetrieval-augmented generation, from basics to agentic retrieval.
BrowseA hands-on guide to running Phoenix locally: trace an agent over OpenTelemetry, then score it with LLM-as-judge and code evaluators.
AgentOps is the practice of running agents in production: observability, evaluation, guardrails, cost governance, and human oversight.
OpenTelemetry in plain terms: the data model, the GenAI semantic conventions, and why building agent tracing on OTel avoids lock-in.