Use-case stack · Pi/ESP32 makers

Home lab automation

Glue sensors, scripts, and LLMs together without sending the whole house to the cloud.

Keep MQTT/automation local. LLMs are optional brains — not the message bus.

Preferred / primary

Self-hosted Automation

Local home automation with optional voice/LLM assist integrations.

In this stack: Device fabric + optional voice.

Details Site

Self-hosted Automation

Self-hostable workflow automation; can call LLMs as steps.

In this stack: Workflows between services and LLMs.

Details Site

Self-hosted Automation

Self-hostable LLM app builder (RAG, agents, workflows) with a polished UI.

In this stack: Self-host RAG/agent portal for the household/lab.

Details Site

Self-hosted Automation

Low-code LLM flow builder, often self-hosted for demos and internal tools.

In this stack: Visual LLM flows for quick automations.

Details Site

Self-hosted LLM / chat

Simplest path to run open models locally (CLI + local API).

In this stack: Local brain for summaries/alerts.

Details Site

Self-hosted LLM / chat

Self-hosted ChatGPT-like UI over Ollama / OpenAI-compatible backends.

In this stack: Chat front-end for lab tools.

Details Site

Web Automation

Cloud automation with AI modules — glue SaaS tools without self-hosting.

In this stack: Cloud SaaS glue when self-host is overkill.

Details Site

Hybrid Coding

Not a chat product — the PlatformIO toolchain paired with an LLM for ESP32/Arduino triage.

In this stack: ESP32 node firmware still needs a real toolchain.

Details Site

Stacks are editorial starting points — swap tools when policy, budget, or hardware demands it.