Rodrigo Hernandez, IoT consultant, educator, and author of Practical IoT Handbook, talks about building LoRaWAN systems that survive outside the lab and deliver real business value. Drawing on his early work with The Things Network and years of hands-on deployments, Rodrigo shares how his journey started with experimental LoRa links and single-channel gateways and evolved into consulting on full-scale IoT systems across multiple industries and countries.

The conversation explores why LoRaWAN is such a strong fit for large, sparsely connected regions like Argentina, and how that same logic applies globally to agriculture, oil and gas, utilities, and building management. Rodrigo explains why LoRaWAN should be treated as a strategic infrastructure layer rather than just a radio protocol, emphasizing long battery life, unattended operation, and the ability to cover remote or difficult environments with minimal operational overhead.

He also digs into the realities of deployment, including why site knowledge still matters, how interference and placement can make or break a project, and what separates successful IoT rollouts from those that struggle. Using real consulting examples, Rodrigo highlights common failure points such as poor sensor choice, lack of on-site expertise, and underestimating the complexity of data handling once devices are live.

The episode closes with a deep look at IoT data visualization and analytics, where Rodrigo explains why clean, well-structured data is essential for meaningful dashboards, how heterogeneous payloads create hidden costs, and why getting data normalization right early is critical for long-term scalability and business insight.

Practical IoT Handbook - Amazon Affiliate Link

Rodrigo Hernandez on LinkedIn

• Helium Global IoT Coverage - Want to know if Helium coverage exists where you need it? Check out this map!
• Helium Foundation - The Helium Foundation's IoT Working Group (IOTWG) has generously provided support for the first 6 months of shows, please go check them out and consider using the Helium LoRaWAN as a primary or backup on your next deployment. With over a quarter million gateways deployed worldwide, it's likely that you have and can use Helium coverage.
• Support The Show - If you'd like to support the MetSci Show financially, here's where you can donate on a one-time or an ongoing basis. Thank you!
• MetSci Show - If you'd like to use our IoT or AI Data Value calculators, or you'd like to contact me, the MetSci Show site is the best way to do it.
• MeteoScientific Console - Use LoRaWAN - The MeteoScientific Console allows you to use LoRaWAN today. As long as you have Helium coverage (and you probably do, about 90% of populated areas in the world have a gateway within 2 miles), you can onboard a sensor. You can always check coverage at https://explorer.helium.com and switch to the "IoT" tab in the top right.

Today's guest

2

on MeteoScientific

The Business of LoRaWAN is Rodrigo

3

Hernandez, an IoT consultant, educator

and author who's been building

4

and teaching real world

5

LoRaWAN systems since the earliest days

of the Things Network.

6

In this conversation, Rodrigo walks

through his hands on path into LoRa

7

and LoRaWAN from building one channel

gateways on Raspberry Pi to helping grow

8

local communities and deploying networks

in challenging remote environments.

9

We talk about why LoRaWAN

is especially well suited for countries

10

with large geographic footprints

and limited connectivity,

11

and how industries like agriculture, oil

and gas

12

and building management

are starting to adopt it at scale.

13

Rodrigo also shares lessons

from consulting projects around the world,

14

including what causes IoT

deployments to fail, why site knowledge

15

still matters, and how data visualization

and normalization becomes critical.

16

Once devices are live and sending data,

this episode is sponsored

17

by the Helium Foundation and is dedicated

to spreading knowledge about LoRaWAN.

18

If you'd like to try Helium, a

publicly available global LoRaWAN for free

19

and support

the show, sign up at metsci.show/console.

20

Now let's dig into the conversation

with Rodrigo Hernandez.

21

Rodrigo, thanks

so much for coming on the show, man. Hi.

22

Thank you very much for inviting me.

23

Yeah, I'm psyched to have you on board.

24

I think maybe we start with this practical

25

IoT handbook because you literally

wrote the book on this thing.

26

Oh, yeah.

27

And that came out. Was it in

May of this year?

28

Yeah, I think it was May or Sean,

I remember really now about the.

29

Yeah, I think it was nine.

30

Yeah.

31

I spoke for

you know, people who want to start

32

experimenting with IoT devices,

microcontrollers, protocols, home

33

automation, sorts of visualization

with Grafana, InfluxDB, etc..

34

So it's like a basic,

you know, training about hardware

35

programing and protocols and platforms.

36

So we cover all but just the first step,

obviously.

37

Yeah. Yeah. It's nice.

38

It seems like a nice intro to.

I've got it on the way.

39

I don't have it in my hands yet.

Otherwise I'd, I'd hold it up.

40

Oh I'm certainly I'm super excited to

to check it out.

41

Great, great.

42

And it's a nice start into IoT

and then eventually into Lora and LoRaWAN.

43

LoRaWAN was my gateway drug into it. Sure.

44

And so

it's cool to see what's going on now.

45

And you've got this much wider view

than my normal guests, who are just

46

usually focused solely on LoRaWAN.

47

You've been doing this consulting with IoT

and AI for the past.

48

Was it five years since 2019.

So six years?

49

Well, I started with LoRaWAN 2017.

50

I started to experiment.

51

I created my first community in my city.

52

Yeah.

53

I was just looking at the emails

I, I had,

54

with, research, Richard Shelby from DTN.

55

Yeah. I mean, I'm going to see him

next two weeks. Yeah. Oh, okay.

56

Yeah, yeah.

57

Because, he wrote me,

when I created, the community

58

and we had, you know, meeting,

I was early

59

you know, I did inspect

because, it was a surprise for me for.

60

I expect to be contacted by these people.

61

I was just starting with that

or where I was on.

62

So it was very exciting

for me to have a meet with him.

63

So, Well,

that was my first experience with, Lora.

64

Or when I started with Lora.

65

Not the one because, you know,

we had this small microcontrollers

66

like Lelo, this,

you know, development course.

67

And I started to experiment with Laura,

just Laura.

68

And then I remembered I built,

69

one channel gateway with a Raspberry Pi,

and,

70

you know, one of these transceivers

with single channel, super high tech.

71

Yeah, yeah, yeah, that was, you know,

72

at that time, it was, very challenging.

73

I remember that, I, I was using, so wide

74

open source software,

they were able to develop,

75

I, I don't remember

who developed the software,

76

but it was some somebody from, university

or something like that, like think.

77

Sure. Probably. You know,

78

and that was my first.

79

But it was very complicated

because I think

80

that you had to compile

the software and deploy it.

81

It was very you know, it's getting easier.

82

Yeah, yeah, but it was complicated. Yeah.

83

So what? I, I need to work several times.

84

For a time, I made,

you know, some experiments in the,

85

I feel I need been deployed,

start this little together

86

in a tower in the city,

with some help from other people.

87

So what?

88

I started the community

there, and it was. And,

89

you know, a few

90

years later, I could officially start

the community with Chuck.

91

I was one on one in my house

and other in some other region.

92

It was very, you know,

you learn a lot doing these things.

93

Yep. So that was my first,

you know, my starting.

94

Yeah.

95

As I said, I learned a lot of things.

96

Well, that when oil started for me.

97

So you start and this is all with TNT.

98

Not to but and stuff.

99

Yeah.

100

You get into it, you're super interested.

101

And then at some point you start saying,

hey, I'm going to start to marry IoT.

102

And at the time I think it was ML

and not quite IoT, and now it's

103

kind of full blown

AI and IoT, and you're consulting on that.

104

What are the typical clients

asking you to do,

105

or what are the opportunities

that you see combining IoT and AI?

106

I started to work with an email recently,

107

but, I know, you know, I been,

108

learning about, email for, if, you know,

for some years I started with there

109

and she calls in Coursera, you know,

110

okay, that was, Yeah.

111

Let's start somewhere.

112

Yeah, yeah.

113

Yes, yes. That was, some years ago.

114

It was a good start to,

115

you know, to understand

the full picture of machine learning.

116

Yeah. So.

117

Well, I have,

I think currently I have, Klein,

118

I'm working on a, prototype

for or for main OCR

119

on, boxes, you know, medical boxes

like this one, for instance.

120

I can show, this one

sometimes you don't have the

121

your this printed but this stamp it.

122

Okay.

123

So it's, it's hard to get the,

the, the letters from there,

124

but so I'm working on this project that

is, for, an important component here.

125

And I, you know,

I been testing for another clients

126

in Singapore, but, currently,

we are not using ML.

127

We we thought, well,

that we don't need it, but, not really.

128

You don't? Yeah.

129

Just taking pictures, but also and but,

130

it was, a good experience

because I have the experience

131

to, you know, to manage this

from the Raspberry Pi.

132

And we are using Pi version five,

133

with, a, a camera, so.

134

Well that we were working with that,

I think, most MRI

135

projects has to do with, ambition,

you know, artificial vision.

136

The.

137

Yeah, that is most of the project I see.

138

I, I see also in Upwork,

139

people who ask for email, typically they,

they ask for a lot of vision.

140

Yeah. So a lot of vision. Okay. Yeah.

141

And then you've been training

if I'm reading this right, you're

142

are you training Argentina's,

automatic control association or you've,

143

you've talked to them or worked with them.

144

Yeah.

145

I'm teach about,

you know, different topics.

146

What do they ask?

147

And how does it interact with LoRaWAN?

148

Do you have to introduce them to the idea?

149

Do they already know about it?

150

Is it kind of mind

151

blowing for them, like walk me through

what that conversation looks like?

152

Well, I think IoT and in particular,

153

you know, LoRaWAN here in Argentina,

we are just starting.

154

We have a, you know, many people

who already know about the tech,

155

but this is you know,

this is a small group of people,

156

you know, techies basically basically

157

people who, you know, or we talk about,

158

know a lot about tech,

but I think but companies,

159

people from other

areas are starting to discover this.

160

We organized,

161

you know, for first time here this year

since conference here in Argentina.

162

And we are starting organizing the next

one for the, you know, the next year.

163

So I, you know, noticed that

there are many people, very interested.

164

They are eager to,

you know, to implement this technology.

165

Sure.

166

I think, obviously I, I love LoRaWAN,

I think is perfect for many cases

167

because, well,

we know all the, the advantages,

168

all the things that you can do,

169

you know, batteries for years, etc.,

remote locations.

170

We have very extensible,

171

a country and the Argentine is huge.

172

Yeah.

173

Well, so you say, but,

174

we have lack of connectivity

in many areas.

175

You go for, you know,

you take a road, it's probably you,

176

you will not have,

177

connection in many times using the trip.

178

So even in the,

you know, fields, farmers, etc..

179

Well, now we have Saturday

night, internet like space.

180

Yeah. Starlink. Yeah.

181

Till we need some kind of,

182

you know, infrastructure

to connect remote locations.

183

Yeah.

184

We are having a boom with, oil and gas.

185

So this industry is, actually,

I have permit now, 3 p.m.

186

here, about the probability project,

about, in the oil

187

and gas industry here

in Argentina for the Patagonia.

188

So, and they are planning to use,

Aurora one.

189

I it's a perfect match, you know, and

these remote locations with LoRaWAN, etc..

190

And so I been asked about,

191

S.O.S buttons, in the road.

192

So we were looking for, Mr.

193

Steak or something like that to,

you know, implement this nature so.

194

Well, I see that there are interest

in this technology.

195

It's interesting

196

because it sounds like it's just starting

to really take off down there.

197

And so you've got the advantage of being

able to look around the world and see that

198

obviously you've been with TTN or you've

you've been involved in for a while.

199

You've seen kind of

they've done all this stuff

200

and you can now say, hey, cool.

201

There's a library of projects

that we can use as demonstrations

202

and implement throughout Argentina.

203

And folks are coming to you.

204

Do you think

205

it would be fair to say

206

that you're one of the central points

207

of people when they're looking

for LoRaWAN in Argentina,

208

or are they generally finding you?

209

Well, maybe they I don't know,

210

but, I'm trying to, you know, to be,

211

Bessemer expert on these things.

212

Yeah.

213

And I've seen you talk about LoRaWAN

as a strategic enterprise asset.

214

So companies are coming to you

and you're saying,

215

hey, don't think of this

just as a radio protocol

216

and some nerdy saying, think of this

as being good for your business.

217

Yeah.

218

How does that

how do you talk to them about that

219

or what do you

what are the points you bring up?

220

Well, one of the main points about Nora,

one is the possibility

221

to, Huberty and attendant devices,

you know.

222

Yeah. In.

223

Yeah, everywhere.

224

No, no matter where they are.

225

So you don't have to go there every week

to see

226

if that works or you are having a problem

with the sensor, etc.

227

it's very robust.

228

Also, it's very easy to deploy.

229

You can configure all the things I,

230

I would like to clarify a point here,

but it's to deploy.

231

But you have to take some, you know,

232

considerations

or maybe, you know, take care of the

233

configurations

234

on the sites

where you are going to deploy the sensors.

235

Because I had this some months ago,

236

I had a not very good experience

with my client in Singapore.

237

Okay. Oh my gosh. Yes. Yeah, yeah.

238

The and the was but it's

okay while this happens.

239

But we configure

I couldn't be there obviously.

240

Sure.

241

And and I did my best, but still

242

when they put all the sensors

243

and the gateway in the field,

some of them didn't work. So.

244

But that is because,

245

we didn't know the, the site, etc..

246

So, I saw, for instance,

I saw a lot of traffic.

247

Nora was in a good way from devices

that were

248

that were not from this, network. So.

249

Interesting. Yeah, yeah.

250

When, by the way, it was the

251

F1 road, this, one competition.

252

Yeah, yeah.

253

So for.

254

Yeah, it was, you know, a project.

255

He had these, trash bins.

256

He had to monitor all these trash bins.

257

Yeah. In the.

258

So yeah, it was a simple project,

but still some things didn't work.

259

I don't know why, because I wasn't there.

260

I think it's still necessary to have,

261

you know, a technician

and that no sound sighted on the side.

262

Yeah. Because, you know, some things

263

you never know.

264

Yeah, yeah, it's one of those things.

265

It's it's, we have a saying it's easy

when you know how.

266

And so it's like.

267

Oh, and you know how you're like,

oh, of course you just got to,

268

you know, change

over the sub band or whatever it is. Yeah.

269

But if you don't know, it's like,

how is the magic working?

270

So when you said F1, I was like,

how is LoRaWAN a fit for F1

271

because it's so fast.

272

But if it's just supporting

273

whatever trash cans, like

you're not putting the things on the cars.

274

Yeah. No no no no no no no. Yeah.

275

And then you've seen

276

a bunch of these IoT projects,

where some of them have difficulties,

277

like the one you just mentioned

and some of them succeed.

278

Are there

characteristics of a strong IoT project

279

where you look at something,

you know, like, oh, that's going to work

280

really well, versus

when we look at something like Ali just

281

they didn't get this one

critical part, right.

282

What are some of those

really good things and really bad things.

283

Well, I

284

think, for instance,

I see a lot of, grow,

285

in the,

building management systems, of course.

286

They, it's a very good, actually,

287

I had, a client in the USA.

288

He was testing an at war in, hotel,

289

you know,

several floors and apartments there.

290

So it was good.

291

It was working very good.

292

You know, stable.

293

The only issues, were,

294

when some gateway

get, offline or something.

295

Sure.

296

That's a common problems,

but when the infrastructure work

297

was working,

it also all goes smoothly on you, sir.

298

Yeah. You have the data? Yeah.

299

So it was a nice experience. Yeah.

300

He finally decided to move everything

to, on premise.

301

We were working

obviously in the cloud, but,

302

it was exactly the same project

that he decided to move on premise.

303

But, about the, LoRaWAN infrastructure

or the sensor.

304

It's a today.

305

It was working very, very good.

306

Everything.

307

On the other hand,

I had a client in Canada.

308

He was working with some sensors

I didn't know,

309

and we had a lot of problems, with the

these sensors, I don't.

310

Yeah, I don't know why, but, sure,

but I sent I.

311

Yeah,

I don't know, so I recommended the, him

312

to buy,

one sensor that they already know.

313

So, we start working with this sensor

that,

314

that, by the way, they are cheap nodes,

315

and they are not expensive at all,

but they work so well.

316

And that then improve a little bit.

317

And then, you know that there

there were some problems,

318

but we didn't know what exactly

what was happening there because it was

319

but we now not that remote location,

but still

320

he had to go to the site to see.

321

Yeah.

322

Because sometimes

you know, people in the,

323

I'm not, I'm,

I'm guessing I'm not sure about what

324

I'm saying, but, I imagine

that maybe some people there, were

325

you know, moving sensors or,

you know, changing something.

326

I don't know, because they were working,

and then they stopped the work.

327

It's.

328

What about your article behavior?

329

So. Okay.

330

Yeah.

331

There are things that you can spray

because you are lacking

332

information from the site, so it's.

333

That part is a tricky.

334

Yeah.

335

Let's wrap this up.

336

We're talking about visualizing IoT data,

which is something that you've

337

you've had a fair amount of experience

with in this Grafana piece.

338

What are things that you look for?

339

Someone comes to you and says, hey,

I want to do this IoT project

340

and I want to see it.

341

How will you go through that with them

and what are really important to you?

342

And kind of common points of,

343

of failure or common mistakes that you see

from new people to IoT visualizations?

344

Well, first you need the, clear strategy

to get the data,

345

transform that data

because you will find, you know,

346

different payloads from many different

manufacturers out there.

347

That is very you know, it's it's

348

very complex to manage

because, you one manufacturers,

349

they use this payload, this other use

maybe this one is not using that.

350

And you know, it's time

stamp in the payload.

351

That is a, you know, something

that you had to solve in some way

352

because you need the timestamp

or maybe the units or etc., etc..

353

So you have to make the data,

354

or more changes so you can manage that.

355

The data effectively, I remember,

one of my projects

356

were for utility company in Germany.

357

And they asked me about

they asked me a specific task that was,

358

you know, they already had

all these sensors distributed in the city,

359

but all the sensors, not all

the sensor were the same kind of sensors.

360

So they had different applications

in the network server.

361

That, by the way, was, DTI.

362

So I had to, you know, move

all the data, transform the data.

363

Send it to I. InfluxDB.

364

In that case,

it was very, you know, complex, but,

365

it was good to, to do this discharge

because,

366

Yeah, it's clean data is.

367

Yeah, it's a big challenge for sure.

368

So it was, good at good project,

369

but very labor intensive.

370

You can.

371

I had to spend some hours.

372

Yeah, I put some cycles into that,

you know, looking through the data,

373

you know, managing.

374

So once you saw a lot of these,

all the data

375

transformation,

then it's easy to visualize and query you.

376

You must look and,

you know, think for one and think about,

377

the query, the

you are going to, to perform the database.

378

So you have to take care of the data

format,

379

measurement feeders, dogs, etcetera.

380

So you can then easily get that data

and show it in the dashboard.

381

That is a this that is a key okay.

382

Because showing data in the dashboard

is as hard

383

as the data is, is, you know, organized.

384

So, is there you see, that is why

you organize this word structure.

385

And then it's easy to, to show it.

386

But you have, mix of different

387

kind of balers, data, fields, etc..

388

What that is, is a big problem

because you had to add up each query.

389

Sometimes you you can do

390

all you have to use very complex queries.

391

So yeah, that is,

you keep your data or machine use, okay.

392

That this is, advice.

393

Yeah.

394

Keep the data motionless I like it. Yeah.

395

Ripping. Dude. Well, thanks

a ton for coming on.

396

Thanks for making the time.

397

I know you're busy,

and I appreciate you coming on

398

and sharing a little bit

about what you know about this IoT

399

data visualization and your low end story.

400

Thanks for having me.

401

Okay. Thank you very much.

402

That's it for this episode

of The Business of LoRaWAN.

403

If you want to go deeper

and actually deploy devices,

404

the MeteoScientific

Console is the fastest way to do that.

405

And honestly, it's

also the best way to support the show.

406

When you use the console, you're not just

listening, you're participating

407

in the same real world LoRaWAN work

we talk about here every week.

408

You can get started with the free trial

at Medio Scientific, Aecom.

409

Huge thanks to the sponsor of the show,

the Helium Foundation,

410

for supporting open LoRaWAN

infrastructure.

411

Alright, check them out at helium Dot

Foundation and at the show

412

has been useful.

413

A quick rating or review on Apple Podcasts

or wherever you listen.

414

This really helps

415

people find it and helps the show grow

so we can help more people.

416

I'm Nik Hawks with MeteoScientific.

417

I'll catch you on the next episode.