wopdiddly retweeted
We published a white paper on what an operating system for physical AI should be.
The short version: a robot only knows what its own sensors can reach. A robot dog on a campus patrol can't know that a door just opened in a building across campus. Connect the door sensor and the robot to the same operating system, and the door opening sends the robot there, turns on the lights and alerts the guard on call.
The video is our patrol-robot demo: an AI agent runs the robot from plain-language orders, and a person decides what happens next.
Read the paper: kiloiot.io/whitepapers/physi…
#PhysicalAI #AIoT #Robotics #IoT #Kilo
Not bad!
CHROME RAIN 🌧️
LEVEL 364 ⚡x4 161,130 PTS
🟨🟨🟦🟦🟪🟪🟪🟪🟪🟪
🟦🟪⬛
BEST STREAK 85
#ChromeRain @waggleandbash
waggleandbash.com/chrome-rai… – at Belfast, Northern Ireland
41
wopdiddly retweeted
The tracker connector supports more than 2,000 vehicle tracker models through one endpoint.
Once connected, a vehicle can appear on the map with its live position and a selected metric reported by the tracker. Operators can also replay its route over a chosen period instead of working from a list of coordinates.
Tracker data uses the same platform as fixed sensors. It can be displayed on dashboards and used by rules and alarms without requiring a separate fleet-data interface inside the deployment.
This gives companies a way to monitor vehicles alongside buildings, equipment and other assets when those operations belong in the same view.
#Chirp #FleetTracking #IoT
wopdiddly retweeted
Back from Amsterdam.
The useful thing about putting a few thousand people who build this stuff in one building is that it makes the real bottleneck obvious, and it isn't the one the industry talks about.
Connectivity is solved. Getting bytes off a sensor stopped being hard years ago. Range, battery life, gateway cost — all of it has been ground down by people who did the work.
What has not been solved is everything after the radio. A payload arrives and means nothing until someone maps it. Automation gets written and then trusted rather than tested. Alerts fire into a void at 3am. And the whole thing still assumes an integrator sits between the hardware and the person who actually cares about the building.
Every one of those is a reason a deployment does not happen. Not a technical ceiling — an effort ceiling.
That's the gap we've spent this year building into, and the conference made us more confident it's the right one.
#Chirp #TheThingsConference #IoT
wopdiddly retweeted
Day one at The Things Conference in Amsterdam.
Our main demonstration starts without physical sensors. We create emulated devices, build dashboards, run rules and trigger alarms, then show how the same configuration can be connected to real hardware later.
We are also demonstrating supported device commands. The user sees the proposed action, approves it and can verify the result against the state reported by the device.
These are important capabilities for commercial deployments because teams can test the operational workflow before a hardware rollout and keep the work when the deployment goes live.
We are here today and tomorrow.
#Chirp #TheThingsConference #LoRaWAN
wopdiddly retweeted
LoRaWAN isn’t going anywhere, and mioty is a very promising protocol. We believe these technologies should work in tandem—and that’s the idea behind this workshop.
@tim_krav_ will be hosting “Build a Mixed LoRaWAN + mioty Monitoring Workflow” at The Things Conference in Amsterdam! 🐦
Tim will walk through how to bring both technologies together in our platform and build a shared monitoring workflow. A practical session to see the platform in action, ask questions, and explore what’s possible when different devices work together.
📅 Tuesday, September 22, 2026
🕐 13:00–13:50
📍 Workshop Room 1 · The Things Conference · Amsterdam
If you’re attending, come join Tim. We’d love to hear what you’re building!
#Chirp #depin #LoRaWAN #mioty #IoT @SuiNetwork
Sui can’t stand out with MEM projects; a true catalyst could be Depin, which is tied to real-world use cases. Take a look at the @ChirpDeWi project in the IoT–Larowan space; we’ve continued to develop our platform—please check it out. We’ll be attending The Things Conference on September 23 at De Kromhouthal in Amsterdam, where our key partnerships will be announced. @ChirpDeWi deserves much more attention.
growth is the only thing that matters to us right now
if your product is bringing new users to #Sui, show us. i want to know what’s working and what we can do to help you make it much bigger
YOU WILL HAVE MY FULL ATTENTION
wopdiddly retweeted
I’ve been waiting over a year to write this post. And next week, I’ll finally get to show you why. 👀
We’re getting ready to release a major Chirp update this week or early next week, and we’ll be showcasing it at the TTN conference.
Our partners have already had a heads-up. Now I can barely keep this one to myself.
For me, this is a huge step forward for Chirp. It opens up a whole new set of possibilities for our IoT platform and makes what we’ve been building much easier to show, explain, and get excited about.
A year of work. Almost time to show you.
Stay tuned. 🔥
wopdiddly retweeted
Chirp alarms can notify several people in sequence until somebody responds.
A team chooses one of five severity levels and builds an escalation chain with recipients, delays and delivery channels. Email, SMS and push notifications can be used at different stages of the same chain.
Schedules and timezone-aware quiet periods control when routine notifications are sent. Duplicate suppression prevents a changing sensor from producing a stream of identical alerts, while repeat behavior can be configured per severity.
The alarm inbox shows what is active, acknowledged and resolved, including the latest trigger time and the values that caused the alarm.
The platform also requires every escalation step to contain a recipient, so a team cannot save a chain that has nobody to notify.
#Chirp #IoT #DePIN
wopdiddly retweeted
The Digital Building Twin was built for more than a single floor plan.
It supports multiple floors, indoor and outdoor areas, and a site footprint traced from an aerial map. The finished model can be anchored to its real GPS position.
The object catalog contains more than 60 ready-made items covering buildings, equipment and outdoor infrastructure. Every object can be connected to a sensor from any supported connector, so different hardware and radio technologies can appear in the same model.
Operators can show live values on the model or hide the numbers and use the color conditions instead. A multi-building or multi-floor deployment can therefore keep its spatial context while the underlying devices continue to use the same dashboards, rules and alarms as the rest of the platform.
#Chirp #DigitalTwin #IoT
wopdiddly retweeted
Chirp can turn a site into a live 3D model and bind sensor data directly to the objects inside it.
A team can model multiple floors and outdoor areas, draw the site in 2D or 3D, or trace the footprint from an aerial map and anchor it to real coordinates. The editor includes more than 60 objects for buildings, equipment and outdoor infrastructure.
Any sensor can be bound to an object. As readings change, the object can change color according to number ranges, text values or true-and-false conditions. Live values can also be pinned in the position where the equipment or monitored area exists.
For a larger site, this gives operators one spatial view of many different systems. They can see which floor, room, parking area or piece of equipment needs attention without translating a list of device names first.
#Chirp #DigitalTwin #DePIN
wopdiddly retweeted
Chirp 3.9 is live — Flight Log. This release removes two things that have capped how big deployments on the network can get: false alarms that make people switch their automations off, and a rules engine that needed one rule per device.
Full changelog and docs: docs.chirpwireless.io/change…
#Chirp #DePIN #IoT
wopdiddly retweeted
Over 2,000 vehicle tracker models work on Chirp, through a single ingest endpoint.
OBD2, CAN, standalone GPS units. You get a URL, you point the tracker at it, and the vehicle becomes an ordinary device on the platform — same dashboards, same rules engine, same alarms as a sensor bolted to a wall.
Which means it inherits everything else without any special-casing. Put it on the map widget with live position and a metric of your choice — speed, fuel, engine state. Play the route back over history. Write a rule against it. Escalate an alert off it.
We think fleets are one of the fastest paths to scale for a network like this, and the reason is unglamorous: it's already someone's job. Fleet operators already track vehicles, already pay for it, and already feel the pain when the data lives in a system that does nothing else. Moving them onto a platform where the vehicle sits next to everything else they monitor is an easy argument to make.
Thousands of devices per customer, and they're already looking.
#Chirp #FleetTracking #IoT
wopdiddly retweeted
For two years AI could talk about the physical world. Now it operates it.
Say "turn on the lamp" and the lamp turns on. Ask it to flip a relay, or change how often a sensor reports, and it does that too.
The part we're proudest of is how careful it is, because something physical is about to happen in someone's building:
It only does what your device is already set up to do. It doesn't invent capabilities.
It shows you exactly what it's about to do and waits for a yes.
And afterwards it checks. A message to a sensor takes a moment, and a sleeping device can miss one — so instead of cheerfully saying "done," it verifies and tells you if it didn't land.
That last one sounds small and isn't. The difference between a demo and infrastructure is what happens when the command fails.
This is what an IoT network is for. Not dashboards you stare at — a physical layer that software can actually reach and move.
#Chirp #PhysicalAI #DePIN
wopdiddly retweeted
You don't need a single sensor to try Chirp anymore.
We shipped the Emulator, and it quietly removes the biggest thing standing between someone hearing about this platform and actually using it: hardware.
Here's how it works. You add an Emulator connector — nothing to fill in, no account to link, no keys to paste. Then you create a sensor and pick a real device model. Chirp fills in the readings that model actually sends: a multi-sensor gives you temperature, humidity, CO2 and air quality, with the right value types and a sensible reporting interval. And then it just starts reporting, on whatever schedule you choose.
From there everything else is real. Dashboards fill in. Automations run. Alerts fire. The rules engine executes against emulated telemetry exactly as it would against a device on a rooftop.
Why this matters more than it sounds: every platform in this space has the same funnel problem. Someone gets interested, hits "you'll need a gateway and a sensor," and disappears for two weeks. Most never come back. That gap has always been the ceiling on how fast an IoT network can actually grow.
Now the evaluation is instant. Model an entire deployment, prove it works, and only then buy hardware — and when the box arrives you point the same device at the real sensor and keep every dashboard, every rule, every alert.
Fewer people bounce. More deployments get built. More of the physical world ends up on the network.
Zero hardware, zero card: app.chirpwireless.io
#Chirp #DePIN #IoT
Here we go!
$chirp is the top daily againer in the entire DePin sector. Took a while for people to wake up to this low cap gem, check out their feed @ChirpDeWi
Device and protocol agnostic, add IOT devices with ease, using AI if you wish. One platform 🔥 – at Tías, Spain
1
97
wopdiddly retweeted
One question decides whether an IoT network grows or stalls: how much of the physical world can it actually ingest?
We rebuilt the whole platform around that. Everything runs on one model — Connector, Connection, Device — and the protocol is just a detail underneath it.
What's plugged into it today:
LoRaWAN, with the network server built in. No external LNS to run or pay for. Gateway registration, joins, uplinks, downlinks, deduplication — all handled.
MQTT, either on our hosted broker or your own. That one line opens up thousands of device models across hundreds of manufacturers.
Zigbee, bridged through MQTT — Aqara, IKEA, Sonoff, Hue, Tuya.
DIY hardware. ESP32, ESP8266, Tasmota.
Vehicle trackers. Over 2,000 models, one ingest endpoint.
The point of the architecture is that adding the next protocol doesn't mean rewriting anything. Devices, rules, dashboards, alarms and access control all sit above the connector layer and don't care what's underneath.
Every protocol we add is more of the physical world that can move onto the network without anyone rebuilding their setup.
#Chirp #IoT #DePIN
What an incredible update, this is the AI IOT project people are sleeping on. Chirp has a protocol and device agnostic IOT platform with AI onboarding that can monitor and control your business, factory, home or smart city. That's what this team is building and you're missing it – at Belfast, Northern Ireland
AI has been able to talk about the physical world for two years. As of today it can operate it. Our Biggest release to date is now LIVE!!
Chirp 3.8 is live — Flight Log
Full changelog and docs: docs.chirpwireless.io/change…
#Chirp #PhysicalAI #AI #DePIN
How many apps do you have to control smart devices in your home? I have one to control an extractor fan in my ensuite, I have one to control a bedroom lights I have another for a smart plug and one for led lights
What utter nonsense, what about that is smart? IOT is fragmented – at Belfast, Northern Ireland
Suddenly your home interacts, controls and monitors itself. Set it on schedules so the lights come on, the blinds open, the heating comes on to wake the kids up for school. One example but countless possibilities. – at Belfast, Northern Ireland
1
1
52
And all the smart devices you have abandoned? I bet you have them but no longer use the apps. Smart device abandonment is common they've become normal devices you manually control. Let Chirp AI manage them for you and make your home AI smart
Chirpwireless.io BRING IT IN – at Belfast, Northern Ireland
1
28
Replying to @wopdiddly @ChirpDeWi
The same goes for Facilities Management, it's worse. The amount governments and commercial enterprise waste having multiple departments and multiple protocols overspending budgets on monitoring with data that is unconsolidated and unshared is the opposite of smart and efficient. – at Belfast, Northern Ireland
1
36




















