Jan-Ole Giebel, founder of J-O. Technik, talks about his rapid journey from early IoT tinkering to building practical LoRaWAN systems for schools and organizations. Beginning with ESP32 sensor experiments in middle school, he quickly ran into the limitations of school Wi-Fi and discovered LoRa—first as simple peer-to-peer radio, then as a full LoRaWAN stack. He shares how supportive teachers and family helped him pursue hardware and programming deeply at a young age, eventually leading him to build CO2-monitoring devices during the pandemic and lead older students in real deployments.

-How early experiments with ESP32s, simple sensors, and Dragino kits introduced him to LoRa and later LoRaWAN’s structured architecture

-The technical challenges he faced with overlapping packets, one-channel gateways, and why LoRaWAN became essential for scaling beyond a few nodes

-The skills he had to develop to make IoT work in the real world, including Linux administration, Python development, virtualization, databases, and managing network servers like ChirpStack

-Why conferences, YouTube, and self-guided learning played a critical role in understanding radio systems, backend servers, and security

-What he sees beginners struggle with most in LoRaWAN and where complexity still creates friction

-His current focus on making IoT practical for everyday users through an application server that hides complexity like payload decoders, device onboarding, EUIs, and downlinks

-How he is integrating LoRaWAN with real-world workflows such as school timetables, automated heating, smart thermostats, and energy reporting

-The type of clients who benefit most from his work, especially schools and organizations aiming to reduce energy costs and carbon footprint without compromising comfort or operational quality

-His perspective on AI tools in development, why he treats them carefully, and where they help versus hinder reliability and security

Jan-Ole on LinkedIn

J-O Technik

Today's guest on MeteoScientific's The Business of LoRaWAN is Jan-Ole Giebel,

the Hamburg-based founder of J-O Technic and the mind behind the Smart Classroom and KliServ-IoT

projects. Jan-Ole started working with sensors and ESP32s back in middle school and quickly ran

into the limits of school Wi-Fi, pushing him into LoRa, then LoRaWAN, and eventually into building

his own back-end platform. In this conversation, you'll hear how he led older students in creating

CO2 monitoring nodes during the pandemic, taught himself Linux, Python, hypervisors, and false

IoT stack administration, and is now turning that experience into a business focused on simple,

energy-efficient automation for schools and organizations. He's part of a new generation

that approaches LoRaWAN from first principles, solving real problems with clear, practical

engineering. This episode is sponsored by the Helium Foundation and is dedicated to spreading

knowledge about LoRaWAN. If you'd like to learn more about the publicly available global LoRaWAN

they run, visit helium.foundation. Now, let's dig into the conversation with Jan-Ole Giebel.

Jan-Ole, thanks so much for coming on. I'm psyched to have you here for the show.

Thanks.

Thanks for having me.

Yeah. It's really cool to see this younger generation, and a huge shout out to Robert

Boggs to introducing us. As we were saying before the show just now, you just got out of high school.

Your English is fantastic. You took 12 straight years of English, and you got into this IT

LoRaWAN thing. Tell me what started it off. How did you find it? What did you need it for?

Basically, I started with IoT back in year seven or eight in school, and my first goal was,

to basically read out some sensor data with an ESP32, and just get into IoT, get into some sensor

techniques, and start with that. And I had a very supportive teacher back in high school,

and the principals were also very supportive, and enabled me to do this kind of stuff in school.

And I finished the rest of the things at home. But basically, the goal was to create an IoT device,

and send the data via Wi-Fi at first. But with the whole school networks and everything like that,

getting into the network and transmitting things like Wi-Fi wasn't as easy as I hoped,

in comparison to home where you have WPA2 authentication and just connect your device.

So that's when I discovered LoRa, not yet LoRaWAN, but just plain LoRa for communication.

And I decided to give that a try.

So I got my first LoRa kit from Draghino with a basic one-channel gateway and didn't really know

that much about LoRa and LoRaWAN at all, and started peer-to-peer communication with a shield

and a node and all that kind of stuff. And then the corona pandemic hit, and we decided to add

a CO2 sensor to the device in order to measure the risk of infection inside the classrooms.

And I was very proud of that. I was very proud of that.

Everyone at the school liked the idea. And so we decided to do that kind of project.

And I was very honored to lead a group of older students back at my old high school.

They were in grade 11 or 12, and I was in year eight. And led them on building this little IoT

device with like an RFM 92 module and an 80 mega something. And we built these devices. And to

add more value to this whole project, I started developing a back-end server written in Python

and the front-end HTML, CSS, JavaScript, and all that kind of stuff to display the measured data.

At first, the server was quite simple, and you could only read out the last transmitted value

and no graphs yet. But then after building more and more end nodes, we discovered that there have

been occurring errors in transmission. So when multiple nodes transmit lower data, they might

overlap, and the receiver got trouble receiving the correct data. So these were just plain data

packages, straight plain strings that were sent and then received and processed.

So then after discovering that kind of problem, I looked into depth, how can I solve this

problem? And that was when LoRaWAN came into play. And at first, this all sounded very complex,

like an application server, like network server, and you need the gateway to connect to all that

kind of stuff. Where do I host it? And how do I get everything integrated, automated,

and all that kind of stuff? So that's why I decided to visit the Things Conference 2022

to learn more about the topic and to meet the persons behind that and everything like that.

And I also learned that I wasn't doing that well with this one-channel Draghino gateway.

So I got my hands on a proper eight-channel or six-channel gateway. Last year, I was lucky enough

to win a curling gateway with all the flagship features. So yeah, that was when I started my

first contact with LoRaWAN. And that's how it started. Gosh,

there's a bunch of questions. It's really nice to see you coming

up and figuring out LoRa. It sounds like as a pretty young kid, if you're an eighth grader,

leading 12th graders, that makes you a super nerd, which is awesome.

Thanks.

I guess for folks who are listening to this and they want to get either

their own kids or just younger people in general into it,

what was it that your teachers did or your parents did for you that made it

super easy for you to kind of do this thing? Can you describe that environment for me?

Basically, my parents gave me the freedom and supported me.

Like financially and with the time and space and freedom that I could approach

these kind of topics. My grandparents were also very, very supportive. We have

still it's an online shop, Conrad Electronics, where you could go into a store. It's like

micro center where you could walk up and grab an Arduino and all that kind of stuff.

So they bought me a Raspberry Pi and all that kind of stuff.

My father is in product management in a company which also focuses on metering and smart metering.

All that kind of stuff. So I had a person to ask for technical difficulties and all that kind of

stuff. But then our paths split and I went with a more computer science direction and

programming direction. So, yes, this is how the environment was for me.

You were saying that IoT has kind of

almost forced you into IT in order to understand what happens. Walk me through some of the skills

that you've needed to learn or the things that you said, Oh, like, I need to have this.

Yeah.

Okay.

and I need to figure this out

in order to get the data from the sensors

to wherever it's going.

Basically, a huge topic for me

was and is the Linux administration.

So you need to be able to set up

like a virtual machine

or if you have a computer laying around,

work with that.

My first computer that I worked with

was a Raspberry Pi.

So I needed to work with the terminal,

work with Python

and keep like a Python app up and running.

Host a web server.

And how do you do that?

High availability wasn't a topic back then

and isn't that much of a topic currently also.

But like security,

what do I need security-wise?

Support for warnings

if I want to host something open to the world

and yeah, all that kind of stuff.

And then later like hypervisors

if I want to host multiple instances

like a database

and my lower WAN server,

ThinkStack, for example, or ShopStack.

How do I split?

How do I connect everything like that?

All that kind of stuff.

And then what were the resources

that you found that helped you?

Was it people?

Was it were you just reading websites?

Was it going to the Things Conference?

At first, I didn't know that many people

that were in IT.

All that kind of stuff evolved

after being in that topic for a while

and meeting more and more people

at the Things Conference

and IT people at the high school I graduated from.

So my first resources were just the internet

and YouTube

and the Internet.

And I bought a Python book back then

for a quick deep dive into Python.

But all that lower WAN stuff,

I learned basically online and from YouTube.

Andreas Spies made some great videos on lower WAN

and so on, yeah.

Oh, that's rad.

Okay, and then what are the,

because you've got such a fresh take on this,

you just got into it,

what are the dumbest things that you see,

I guess, folks in lower WAN doing where you're like,

oh man, this doesn't make sense at all.

Why didn't they build it this other way?

That's a great question.

I don't currently have an example in mind,

but maybe if you want to approach P2P communication,

like there are, like with MeshTastic, for example,

there are great projects with P2P communication,

but maybe if you want to connect a sensor to something,

don't try to use a slash as a separator

for transmitting data and use multiple nodes with one channel.

I think that's a good example for communication

when wanting to build a network with multiple devices,

like something like that.

It's always good when you come in new to anything.

You've got this perspective that people who've been in it

for a long time don't have,

and usually there's some really good insight there.

I remember getting super frustrated early on at joins

and not being able to figure out,

like there wasn't a clear path to understanding

why something wasn't working,

and that still seems to be the case

as you're doing this thing.

This is kind of the number one thing I see

with people new to LoRaWAN is like,

all right, I think it's working, but why isn't it working?

And there's no step-by-step, steps one through 20,

like check this, check this, check this, check this.

Yeah, yeah.

Cool, all right, well, let's talk business.

That's the name of the podcast is The Business of LoRaWAN,

so we might as well talk about it.

You're starting this business now.

Walk me through where you are

and where you want to go in the next year or so.

So I'm currently focused on providing IT and IoT services.

So if you,

like an IoT stack provided and hosted,

I can help you with that.

And a big part of what I want to achieve

is still a product project,

but I'm hoping to integrate that into my business as well.

So my application server that I developed

over the past couple of years,

and my goal with that is to provide a solution

for an end customer who doesn't know that much,

of the whole IoT topic.

So who doesn't really want to get into

what is an application server,

what is LoRaWAN,

and what do I need to connect

in order to be able to have a beautiful dashboard

in order to reach data or to get reports

and all kinds of that stuff.

So I wanted to do it as simple as possible

for end customers to just say,

hey, I want to have a smart heated office.

Like what do I need in order to do that?

And how can I do that?

And that's where I want to come into play

and basically say, yeah, I've got an application for you.

You don't need any knowledge on IoT at all.

You just get the gateway from me

and you just upload the file

and every little details like UIs

and all that kind of stuff are buried into the files

and you just add them

and you don't need to know like what is a payload decoder,

all that kind of stuff happens in a device repository

inside the server.

And I'm also working on a public device repository

with communities called Smart Digital Regionale

here in Germany,

where we are trying to build a device repository

for LoRaWAN, but possibly also for other communication standards

where the devices are described

and every input and output like uplink and downlinks possible,

uplinks and downlinks,

and payload decoders are all described into there.

And this is also what I think my application makes special

or where I want to come into play is like,

yeah, okay, you don't need to have any knowledge at all.

You just say you want integration, for example,

into a school timetable.

It's a lot of work to do time automations.

Like if you've got a whole class and a whole school

and need to reschedule a lesson,

for example, doing that in a IoT server currently

is a lot of work.

And I don't know of any IoT server

who would allow me to do that easily.

That's why I have an integration with the timetable software

where the principal just needs to reschedule one lesson

and doesn't need to know anything about IoT,

but the classroom isn't heated anymore, for example.

Got it.

Oh, that's smart.

And that doesn't exist right now.

That's cool.

Yeah, and all these little things,

I want to make them as easy as possible for end user.

I'm optimizing my software more and more,

and there isn't like the final 100% production ready software yet.

I'm currently in the development and it's running in my old school.

And yeah, so far it's working relatively great.

There are still some bugs and errors that I have to iron out.

But in general, there was a lot of interest in this software.

And local media and press attended the press conference we held

in order to tell the world about this smart classroom, as we call it.

Yeah, and that's what I want to achieve with my company in the long run,

to do IoT simply for small, medium, and at one day, large customers.

Very cool.

Let's see, I've got two more questions.

The first one is, what does a great client look like?

Like, what is your next client that you want to find?

And secure look like?

I think the perfect client might be one where it's where we can build a strategy

to improve energy efficiency and energy efficiency,

carbon footprint and all that kind of stuff.

Not necessarily pretty easily, but yeah,

focus on implementing the IoT solutions into their workflow as easily as possible.

Like take, for example,

a school and they want to have their timetable automated.

They want to have their smart thermostats automated.

They want to do readings on their energy meter, on their gas meters.

They want to get reports for all that kind of stuff,

like how much energy did I consume over the last couple of months?

How can I improve on that?

Like, how can I save money on the one hand,

and keep the carbon footprint down, save energy,

but do all that kind of stuff without losing any quality

of the actual work that is being done at the place?

Yeah, you can turn the energy off,

but if the classroom's freezing cold or no one's in it,

it doesn't matter how much money you're saving.

Okay.

And then the second question is use of AI,

I used to think this idea of a device repository when I first started,

and I haven't started that long ago, probably,

which is after you, 2021, 2020, something like that,

would be super nice.

Like, oh, if I could just put a device repository

in the MetSci console that I run that's a public console,

that would be great for people.

But what I've seen is that AI,

specifically GPT or Cursor,

will just build the thing for you.

How much are you using AI in what you're doing,

whether it's LLMs or something else?

Are you touching that?

Do you trust that?

Do you use it?

What's that look like?

Trusting?

I wouldn't say so.

Basically, I think AI can be a starting point

if you have no idea for a specific topic,

but you have to take the output very, very carefully,

especially when it comes to IT security

and all that kind of stuff.

There can be huge vulnerabilities

and you need to understand the output.

You definitely need to do that

in order to not ship a product

with many security vulnerabilities

and features that won't work.

And I think you can use AI

for repetitive jobs,

for repetitive coding,

or to build structures

or find errors

that you don't find in your code.

But I had many experiences

where I found the use of AI very frustrating.

I basically told it,

yeah, I want that feature,

or please implement me that,

and this,

and it failed on doing that multiple times.

And I've got error after error after error.

Sure.

And it was not able,

to fix the error.

Then I had to go into it manually

and find the error myself.

And then I thought,

why weren't you able to get that

but program simple things?

And that's also why I think

that our jobs are safe for now

because that are things

that I can't do right now.

And yeah,

I'm not sure when they will be able to do that.

Okay.

Super cool.

Jan-Ole, thanks so much for making the time today.

I really appreciate you carving time out of your day.

I know you're starting a business.

This is a very busy enterprise.

So thank you for cutting this out.

Yeah.

Thanks for having me.

That's it for this episode of The Business of LoRaWAN.

I built this for you,

the one person in about 100,000

who actually has an interest

in how this tiny little slice of the world works.

Of course, this isn't just about you and me.

It's about everyone in LoRaWAN

and how we can work together

to make an exceptional thing.

LoRaWAN is a dispersed community

with little pockets of knowledge all around the world.

And most of them don't talk to each other

as much as I'd like.

So the first,

and best thing we can do to make this show better

is to get more guests on

who I don't even know exist.

I want to talk to strangers.

Strangers who are your friends.

Please introduce me to the most rad LoRaWAN-er you know,

or point them my way,

or help reach out and give me a name.

When it comes to running down LoRaWAN guests,

I can track a falcon on a cloudy day.

If you can remember metsci.show,

you can find me.

That's M-E-T-S-C-I dot S-H-O-W.

Metsci.show.

Metsci.show.

Okay, so sharing knowledge by getting great guests on

is the first and by far the most important thing

we can do to make this better.

The next best thing for the show to do is the usual stuff.

Subscribe to the show, give it a review,

share it in your corner of the world.

Again, that's Metsci.show.

Finally, if you want to support the show financially,

you can do that over at support.metsci.show.

You'll see options there for one-time donations

if you really like this show,

as well as an ongoing subscription option

if you think this show is worth supporting.

So if you're interested in supporting the show,

for the long term.

If you want to try LoRaWAN for yourself,

sign up for a MeteoScientific account

at console.meteoscientific.com

and get your first 400 data credits for free.

That's enough to run a sensor for about a year

if you're firing every hour.

The show is supported by a grant from the Helium Foundation

and produced by Gristle King Inc.

I'm Nick Hawks.

I'll see you on the next show.