v0.1.0 · open source · MIT

ingest. validate.
persist. repeat.

Move data from your sensors to your database — without writing a single line of glue code.

Connect any broker, describe your data in a visual editor, and stream it clean to where you need it. Self-hosted, runs on a single binary. Yours.

▸ Try the demoview on GitHub →
time to first stream
~5 min
from install to data flowing
code you write
0 lines
everything from the web UI
where your data lives
your DB
never ours · no SaaS, no telemetry
▸ Try the demo
// how it works

Three steps, no code.

Your devices
send messages
RabbitMQ · MQTT · Kafka
↓
b
iot bees
cleans & routes
validation · transforms · replicas
↓
Your database
ready to query
InfluxDB · local log
// the workflow

Four screens from broker to dashboard.

Each step is a single screen in the web app. No YAML to write, no rebuild, no restart.

01

Connect your broker

Pick the type, fill three fields, save. iot bees remembers it.

// TYPE
RabbitMQ▾
// NAME
Temperature sensors – plant A
// QUEUE
sensor.temperature
02

Describe your data

List the fields you expect and the valid range. Bad messages get dropped, good ones get cleaned.

temperature
FLOATrequired
-50 → 150 °C
humidity
FLOATrequired
0 → 100 %
sensor_id
STRINGrequired
tag
+ add field
03

Wire a pipeline

A pipeline links one broker, one schema and one store. Pick from dropdowns, hit save, you have a working data path.

// SOURCE
Sensors plant A
// SCHEMA
Temp + humidity
// STORE
Influx (prod)
04

Start it and watch

Press start; iot bees spins up workers and starts streaming. Stop, scale up or tear down whenever — no restart needed.

// THROUGHPUT
3.1k msg/s
RUNNING
// use cases

Built for the messy reality of the field.

If your data sits between sensors and a database — and you would rather not write a custom service for every project — iot bees fits.

// industrial

Factory floor

Watch temperature, vibration and pressure live. Alert when something is off, keep a clean history for shift reports.

// agriculture

Smart farming

Soil moisture, humidity and EC from field stations, all in one place. Normalize the units, ignore the broken stations.

// energy

Power monitoring

Meter readings cleaned and converted to kWh on the fly. Store raw and computed for the audit you didn't know you'd need.

// buildings

Building automation

Centralize HVAC, occupancy and air quality from many buildings. Drop bad packets, forward clean records to your analytics.

// architecture

Actor-based, fault isolated, scalable.

Each pipeline is supervised independently. If a single replica crashes, only that replica is restarted — the rest keeps streaming. Built with Actix on Tokio.

SystemActorSupervisor
  └─ PipelineSupervisor                    (one per pipeline)
        ├─ Replica 1: [consumer] → [processor] → [store]
        ├─ Replica 2: [consumer] → [processor] → [store]
        └─ Replica N: [consumer] → [processor] → [store]
// self-host

Yours, on your hardware.

Your data, always
Nothing leaves your network. No SaaS, no telemetry, no third-party cloud in the path.
One binary, SQLite
No Docker required, no orchestrator. Runs on a NUC, a Pi, a bare-metal server — whatever you already have.
MIT licensed
Fork it, audit it, ship it. No vendor lock-in, no expiring license file.
▸For your DevOps team — three commands
$ git clone https://github.com/manuelmj/iot-bee.git
$ sqlx migrate run --database-url sqlite://data/iot-bee.db
$ JWT_SECRET=change-me make run