Meet me @ Kranky Geek San Francisco 2018

November 5, 2018

Kranky Geek is happening this year again, the date is Nov 16, and we’ve got the best lineup of speakers for you.

Kranky Geek started almost by mistake. Like most good things that happened to me. It wasn’t planned. The result though is becoming a tradition by now, where I get to work with Chris Koehncke and Chad Hart for a period of time that can be considered quite intense (we’re all too opinionated).

Google, along with our other sponsors make this event happen. We only curate the content to make sure the end result is great.

In last year’s event, we started looking at the domain of AI. You can find the recordings of that event on YouTube. The feedback we got was positive, so this year we’re taking a step further here. Many of the sessions will focus on machine learning and AI and its impact on real time communications.

What’s on the Agenda?

AI in RTC.

As always, our intent here is to focus as much as possible on services and applications that are running in production already. It won’t be theories about what can be done but what are people doing. Today.

The updated agenda can be found online. It might change a bit in its ordering, but it is mostly ready.

This year, we have some brand new speakers for you:

  • Discord will be giving a session about their service and what they had to do with WebRTC to make it work for their use case. My suggestion? Read their post to get ready for this session – it will be really interesting
  • Houseparty are joining us for the first time as well. Tinkering with machine learning on device. One of the main challenges these days is deciding where to run inference with machine learning – on device or in the cloud. We will see both options throughout the day
  • Agora will explain what they are doing to improve video quality in real time on mobile devices by using machine learning
  • Voicera will be talking about the challenges in speech recognition when it comes to handling meetings
  • Dialpad are there to talk custom vocabularies. Every company has that. How do you transcribe Kranky Geek? That’s a question I’ll ask in the Q&A of this session…
  • Intel will discuss newly open sourced visual processing tools to help you build out your application
  • RingCentral is joining us late in the game. We’re figuring out with them a stellar topic for the event

We also have some “repeat” speakers:

  • Facebook this year will give us a sneak peek at the technology (and AI) behind their new Facebook Portal device. What I am really keen on hearing is what decisions they made to get their “follow you around” feature to work
  • Voicebase will focus on paralinguistics this time. The nuances of speech that aren’t text – and how to capture their meaning
  • Callstats will be discussing this time the use of looking at ongoing call data using… machine learning
  • IBM will be all over voicebots and their uses in contact centers. We will get to look under the hood on how these get implemented
  • Nexmo are going to show us the complexity of connecting real time voice streams to cloud based speech to text engines. (technically, there are a new speaker, but I figured that now that TokBox is part of Vonage which also owns Nexmo, they are repeat speakers)
  • Google will give an update on Chrome’s implementation of WebRTC, with a focus on 1.0. They will also give a deep-dive into the upcoming architectural changes in Chrome’s audio processing engine
  • Microsoft is going to give us a demo of WebRTC, Mixed/Augmented Reality and HoloLens. And we’re saving this for last so you’ll stick around

We are expanding our family of Kranky Geek speakers and Kranky Geek companies, which is a true joy. I can’t wait to hear your feedback once the day is over.

Our sponsors this year

As always, the event is practically free to attend (there’s a $10 admission fee that gets donated to Girl Develop It).

The companies that made this event happen this year are Google, Intel, Agora.io and Nexmo who are our premium partners for the event; Callstats.io ,Voicebase and RingCentral who are our silver partners for the event.

No fire drill 🙂

I am not sure if this is good or bad. We had a surprise fire drill last year. We knew about it about a week or two before the event. It cause so much headache for us. And a lot of worries.

It ended up pretty well, with our audience and speakers getting a one hour break outside on a beautiful sunny day. Almost all of them came back after the drill, which isn’t obvious or even expected.

Many were happy for the break – and the smalltalk that ensued during it.

Hopefully, there will only be pleasant surprises this year as well.

What are we looking for in Kranky Geek?

We had to turn down a few vendors who wanted to speak. This is a process that takes place every year.

There’s no specific set of rules of what we approve or don’t as a session in Kranky Geek, but for me it boils down to this:

  1. Something new that wasn’t discussed at Kranky Geek before
  2. Preference to something running in production at scale
  3. An interesting topic that would appeal developers
  4. Related to real time communications
  5. A speaker that can “hold a room”

While the lineup of speakers for this year is full, if you want to speak in future Kranky Geek events – be sure to catch me during the event for a chat.

Should you travel just for this single day?

I got this question a few times in the past few weeks.

My guess is that if this is the only thing you’re doing in San Francisco and coming for, then skip it. Especially if you are traveling from abroad.

That said, if you want to feel where WebRTC is headed, talk to many of the people who deal with it daily in the real world, then this is the place to be. So many discussions take place during the breaks that it might be worth coming only for the breaks… I know a person or two that are coming only for that.

We try to make Kranky Geek special and unique. We work hard to select the speakers and work with them on their presentations. All to make it worth your travel, wherever you come from.

Can non-developers attend?

We received this question recently.

There is no easy answer to this one. On one hand, the event and its session are technical in nature as our focus is developers. On the other hand, the sessions are short (20 minutes all-in-all), so our speakers tend to focus on the essence and not dive too deep into the nitty gritty details. So a tough call.

My suggestion? Check out some of the session recordings on YouTube from past events and make your decision based on that.

Register now

Yes. there’s this minor detail.

You need to register to attend. There’s limited room capacity, and at some point, we will need to close the registration.

We’re already half full in our registration list, so save your spot now and don’t wait.

Do you want to meet me prior to the event?

I’ll be in San Francisco Nov 12-17. Nov 15-16 are reserved for Kranky Geek. The rest for meetings with people – around WebRTC, CPaaS, testRTC, my WebRTC course, consulting and just catching up.

If you want to meet me during that week, leave me a note.


You may also like

Comment

Your email address will not be published. Required fields are marked

  1. I live in San Jose but will only be up for the day. No need to meet with you per se, but there is one very nagging issue I have with mobile iOS and the lack of implementation, not only in mobile Safari/ mobile ios PWA (which is really frustrating and synonymous to Apple) but also all the other iOS browsers like Chrome, Firefox etc and you may be able to connect me to the proper decision makers I normally don’t have access to.

    Any and all of these mobile Safari competing browsers are and have been capable of implementing an interface to getUserMedia (specifically for audio and raw without filters by default) as well as audio analyzer nodes for real time FFT from the microphone. Even before Apple announced acceptance of webRTC. But they haven’t.

    I would like to argue the case for them doing so now and hopefully give them a reason to do it. But I don’t know who to plead my case to and doing it in person would have a much bigger impact.

    And second, does no one else see the problem with Apple holding the keys to being the only access to creating a PWA invokable launch icon on an iPhone? Sure Chrome or Firefox COULD give me real time access to the mobile microphone and HW / SW assisted FFT for my web app, but that is not very competitive if mobile Safari is the only Mobile PWA builder in town.

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}