Blog

News
& Guides.

Product launches, architecture deep dives, and how-to guides from the team building the AI workforce platform.

EngineeringFebruary 27, 2026

How to Build an AI Coding Agent That Actually Ships

Most AI coding tools generate code. Very few ship it. Here's the architecture behind an AI agent that writes, tests, deploys, and monitors production software autonomously.

Deep DiveFebruary 27, 2026

260 MCP Tools: The Unfair Advantage Over Copilot and Cursor

Model Context Protocol gives AI agents access to 260+ tools — databases, APIs, cloud providers, browsers, and more. Here's why MCP changes everything for agentic coding.

AnalysisFebruary 27, 2026

The Real Cost of AI Coding Agents in 2026

Devin costs $500/mo. Cursor costs $20/mo. Claude Code costs $20/mo. We broke down every AI coding agent's real cost — including the hidden fees nobody talks about.

TutorialFebruary 27, 2026

Deploy an AI Marketing Agent on Discord, Slack, and Telegram

Step-by-step guide to deploying an AI marketing agent across Discord, Slack, and Telegram. One bot, persistent memory, every channel — in under 10 minutes.

GuideFebruary 27, 2026

AI Content Automation: From Blog Posts to Ad Copy in One Agent

One AI agent that writes blog posts, generates social media content, creates ad copy, writes email sequences, and optimizes for SEO — all from a single prompt.

AnalysisFebruary 27, 2026

The $50K Marketing Stack vs One AI Agent

We analyzed what companies actually spend on marketing tools and compared it to running a single AI marketing agent. The savings are staggering — $75K+/yr.

Deep DiveFebruary 26, 2026

Hanzo Bot vs Every AI Coding Agent

Cursor costs $20/mo. Claude Code costs $20/mo. Copilot costs $19/mo. Devin costs $500/mo. Hanzo Bot starts free and does more at $15/mo. Here's the full breakdown.

GuideFebruary 26, 2026

AI Marketing Agents: Automate Your Growth Stack

Stop paying $50K/yr for marketing tools. Deploy an AI marketing agent that handles content, campaigns, analytics, and engagement across every channel for $15/mo.

LaunchFebruary 26, 2026

Crypto Skills: Agents Go Onchain

Pay, trade, build apps — no human in the loop. Agents can now use crypto autonomously with USDC on Base via x402.

ArchitectureFebruary 24, 2026

The Vertically Integrated AI Cloud

What every agentic harness actually needs — and why owning the full stack from LLM gateway to compute tiers changes everything.

AnnouncementFebruary 20, 2026

Introducing Hanzo Bot

Your AI team in a box — one bot, every role, every channel. Deploy autonomous AI agents that run locally or in the cloud.

Origin StoryFebruary 20, 2026

The Road to Hanzo Bot: Ten Years of Building AI Infrastructure

From the first Hanzo API in 2014 to Jin hypermodal research, Zen MoDE models, Enso diffusion, and a 260-repo AI platform — the full story of how Hanzo Bot was born.

LaunchFebruary 9, 2026

Zen4: Open Abliterated Models from 4B to 1T

The Zen4 family brings abliterated open-weight models from 4B to 1T+ parameters, removing refusal bias while preserving full capability.

LaunchSeptember 15, 2025

Zen Pro: Professional 8B AI with Instruct, Thinking, and Agent Modes

Zen Pro is an 8B professional model with three specialized variants for chat, reasoning, and tool use — runs on a single 16GB GPU.

ResearchSeptember 15, 2025

Zen MoDE: 47 Models, One Architecture

Inside Zen — Mixture of Diverse Experts (MoDE) architecture that activates only the parameters it needs. 47 models spanning text, code, vision, audio, video, 3D, and agents.

LaunchJune 15, 2025

Zen Max: Frontier Reasoning at 671B

Zen Max is a 671B MoE reasoning model achieving AIME 99.1%, SWE-Bench 71.3%, and 200-300 sequential tool calls without human intervention.

LaunchMay 1, 2025

Zen Coder: Code AI from 4B to 480B

Zen Coder is a family of code-specialized models from edge to frontier with 128K context, 512K thinking tokens, and MCP integration.

LaunchMarch 15, 2025

Agent SDK: Multi-Agent Orchestration with 260+ MCP Tools

How we built a Go-based agent SDK that coordinates multiple AI agents with shared memory, tool delegation, and 260+ MCP tools — the orchestration layer behind Hanzo Bot.

LaunchMarch 20, 2025

Operative: Giving AI Eyes and Hands

We built a full computer-use stack — Operative, CUA, ACI, and Overlord — so AI agents can see screens, click buttons, type text, and automate browsers. Open source.

LaunchFebruary 15, 2025

Zen Nano: 0.6B Edge AI Model

Zen Nano is a 0.6B on-device AI model built for mobile and embedded deployment, achieving 44K tokens/sec on M3 Max with a 40K context window.

ResearchApril 1, 2024

Enso: From Diffusion LLM to AI-Native Browser

Enso started as a dLLM — a diffusion language model for text generation. It evolved into an AI-native browser with diffusion intelligence built into the rendering engine.

ResearchJuly 15, 2023

Jin: Our Hypermodal AI Architecture

Inside Jin — a unified hypermodal architecture built on JEPA principles that processes text, images, audio, video, and 3D through shared representations. Not multimodal. Hypermodal.

ArchitectureAugust 1, 2023

LLM Gateway: 391 Models, One API

Why we built a unified LLM gateway that routes to 391 models — 47 Zen and 344 third-party — through one API. Automatic fallback, cost tracking, and budget controls built in.

Build with Hanzo

Deploy AI agents that code, market, and operate — starting free.

Get Started Read the Docs

News& Guides.