Engineering Blog

Engineering blog

Deep dives into agent architecture, memory systems, and self-hosted infrastructure.

July 7, 2026

Release Notes: v0.25 – v0.28

Four releases in one week: GitHub webhook integration, per-agent LLM overrides, token usage dashboard, and chain-of-thought feedback.

Read →

July 3, 2026

How I Replaced My Cloud AI Assistant with a Self-Hosted Agent

A personal account of migrating from cloud AI assistants to humux — what worked, what broke, and what I'd do differently.

Read →

June 28, 2026

Subagent Patterns: Scoped Delegation for Complex Tasks

How to spawn scoped sub-tasks with inherit-never-widen security, budget controls, and sync or background execution.

Read →

June 23, 2026

Group Chat Engineering: Multiple Agents, Zero Loops

The addressing protocol, reply decision gate, rate cap, and speaker tags that make multi-agent Telegram groups work.

Read →

June 18, 2026

Why Multiple Agents Beat One Generalist

Specialized agents with different skills, tools, and personalities outperform one monolithic agent trying to do everything.

Read →

June 11, 2026

Four Ways to Use humux: Detailed Conversation Walkthroughs

Real conversation flows for four personas — from code review to email triage to multi-agent group chats.

Read →

June 4, 2026

Voice Pipelines for Agents: On-Device STT and TTS

Building offline-capable voice interaction with faster-whisper for speech-to-text and Kokoro 82M for text-to-speech.

Read →

May 28, 2026

Secrets Management for AI Agents: Why .env Files Aren't Enough

How a two-tier encrypted vault keeps credentials out of the model context and why .env files are a security risk for autonomous agents.

Read →

May 19, 2026

Permission Engineering for Autonomous Agents

Glob-pattern rules, approval flows via Telegram, and vault-sealed secrets that never enter the model context.

Read →

May 5, 2026

Memory Architecture for Long-Running Agents

Four tiers of memory: lexical retrieval, semantic search, importance-based forgetting, and hygiene via deduplication.

Read →

April 21, 2026

Skills Over Code: Teaching Agents via Markdown

Why describing CLI tools in markdown files beats writing Python adapter classes for every new agent capability.

Read →

April 2, 2026

Running a Full Agent Stack in a Single Container

Architecture decisions that make it possible to run LLM orchestration, voice, email, calendar, and admin UI in one Docker container.

Read →

March 17, 2026

Designing Multi-Channel Agent Architectures

How to build agents that operate across Telegram, email, calendar, and CLI as a unified system with a common message abstraction.

Read →