🧠All Things AI — by Subhojit DeyAll Things AI
🌱Start Here🔧Build with AIDaily StackDevelopersVibe CodingOthersLocal🏢Industry🛡️Legal🔬Deep Dive📰News
🧠 All Things AI
🌱🧠🔧⚡⚡🤖✨🔍🔶🎯💜⚡🪟🦙🤗🦞🔁🌊✕🔀🛠️🏢🛡️✅🏭🔬📰
Build with AI
🔧Build with AI
Chatbots
RAG
Agents
Workflows & Automation
Voice Assistants
Evaluation & Testing
Computer Use Agents
Reference Architectures
Model Economics
Knowledge Graphs
⚡Make AI Work
Create Deliverables
Software Development
Data & Database Work
Backend Engineering
Frontend & UI/UX
Personal Productivity
AI Strategy & Product
Build with AI
🔧Build with AI
Chatbots
RAG
Agents
Workflows & Automation
Voice Assistants
Evaluation & Testing
Computer Use Agents
Reference Architectures
Model Economics
Knowledge Graphs
⚡Make AI Work
Create Deliverables
Software Development
Data & Database Work
Backend Engineering
Frontend & UI/UX
Personal Productivity
AI Strategy & Product
Build with AIChatbots

Chatbots

Chatbots are the most common AI product pattern — but building one that is actually useful requires understanding the design decisions behind it. This section covers the different types of chatbots, how to manage state and memory across turns, when to retrieve vs generate, and how to measure whether your chatbot is working.

In This Section

FAQ vs Task vs Copilot Bots

The three chatbot archetypes, their different architectures, and how to choose the right one for your use case.

State & Memory Management

How to maintain context across a conversation — short-term window management, long-term memory, and summarization strategies.

Retrieval vs Generation

When to retrieve a pre-written answer vs generate one — and the hybrid patterns that combine both for best results.

Chatbot Evaluation

How to measure whether your chatbot is actually working — metrics, test sets, and evaluation without a ground truth label for every response.

Previous← Build with AINextFAQ vs Task vs Copilot Bots →

Page built: 01 Jun 2026