Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: Chatbots
Five LangChain Memory Strategies, From Buffer to Entity Memory
How chat models forget between calls and the five memory patterns for LangChain and LangGraph: buffer, window, summary, summary buffer and entity, with trade-offs.
7 min readThread Memory for a LangGraph Support Agent With SQLite Checkpoints
How a LangGraph support agent keeps multi-turn context using SqliteSaver and thread ids, and why short-term thread memory differs from long-term memory.
6 min read13 Claude Habits to Fix: Prompts for Agreeable, Confident, Forgetful Chats
Thirteen Claude habits nobody warns you about, from agreeing too fast to sounding sure when it's wrong, each paired with a short prompt that corrects it.
7 min readDesigning Our Multi-Agent Online Shopping Platform: Why One
Why one big chatbot failed for online shopping and a multi-agent LLM platform worked: the god-prompt trap, math and logic errors, prompt bloat and latency.
13 min readChromaDB as the Vector Brain for Doc-Aware Chatbots
Why RAG chatbots need a vector database, how ChromaDB collections and metadata filters work, and when to prototype on Chroma before moving to something bigger.
6 min readRAG Explained: Stop Chatbots From Inventing Company Facts
A practical walkthrough of the RAG pipeline—loaders, chunking, embeddings, vector stores, re-ranking, hybrid search, and RRF—plus when not to use retrieval at all.
8 min readChat Interface or Agent Loop? Deciding What Your AI Feature Needs
Learn how to tell a conversational chatbot from a goal-driven agent, what the agent loop adds, and how to decide which one your AI feature actually needs.
12 min readChatbot vs AI Agent: What Actually Separates Them Beyond the LLM
Learn why the real difference between chatbots and AI agents lies in the surrounding system architecture—tools, planning, and actions—not the LLM itself.
6 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.