O3 BARISTA INNOVATION DAY TALK

The following is a transcript of the talk I gave during Philly Tech Week 2017’s Innovation Day event at o3 World.

Thanks for looking at these slides! And many thanks to O3 World client and sponsor for coffee related happenings, La Colombe!

I’m Howie Ross and I’ve the Development Director at O3 where I strategize, architect and lead a team of some of the most talented devs in the business.

O3 is a digital product design and development agency. We work with companies to create tools that — hopefully — transform a market opportunities into revenue.

Traditionally, we’re probably like a lot of digital agencies out there in that our solutions existed primarily on a screen. As we’ve grown into the product market, and with the rise of IoT, it became clear that to better serve our clients, we needed to extend those solutions into the real world and physical space.

We have a wide range of clients across industries. However, we do have an emphasis on FinTech. We also do enterprise websites and ecommerce.

We are all about using the right tool for the job. We have a full stack Dev team — from devOps and back end to JS engineering and front end. Our devs play a role in strategy, design, and development. And our designers are familiar with dev in order to understand feasibility, capabilities and trends in the industry.

So, over 2 years ago we started O3 Labs with the simple goal of experimenting with emerging technology to add to our arsenal of services. Based on that criterion, we’ve been very successful. We’ve learned a lot, people love the experiments and the team is very engaged with the program overall. It’s helped us from a marketing and recruiting standpoint as well.

Labs gives us space to explore opportunities for technology within our clients’ markets and beyond. It aligns with O3’s core value of “Always Growing”. It gives us an opportunity to play with new or just interesting tech that may be outside of our current core competency and skill set.

We usually start a new lab with an ideation session around a problem or technology. Once we’ve converged on a specific lab we want to pursue, we invest about 40 hours in a Proof Of Concept. If we feel that the POC has potential, we’ll invest in polishing it for public use and open source. This structure forces us to ship, document, market and move on to next lab.

Some of the technologies we’ve experimented with include Near-field communication (NFC) tech like iBeacons, AR, AI, Machine learning, screenless UI, voice control, my new favorite blockchain and Ethereum…and numerous other buzz words…

Some of the specific labs we’ve done include Theme Music (see video).

Another one of our labs is RoomBot, a bot that lives in our conference rooms and is connected to our Google Calendars. It politely lets people know when they’ve overstayed their welcome by Making an announcement and Dimming the lights via Philips Hue Smart Bulbs.

And then there’s Project Mercury or as the team affectionately refers to it: O3 Poops! Poops is a macOS menu bar app that’s connected to a Node server via a persistent SocketIO connection which is connected to wifi enabled RaspberryPIs in the bathrooms which are hooked up to alarm system window / door sensors. The assumption is that if the door’s closed it’s in use and you don’t even have to get up to check.

It solved a legit problem and we’ve quite a bit of interest from other businesses with the same problem.

But this is actually about O3 Barista. Barista is a web app that analyzes a twitter feed in order to recommend a coffee roast. It uses the IBM Watson Personality Insights API to use linguistic analytics and personality theory to infer attributes from a person’s unstructured text.

Personality Insights outputs a tree of cognitive and social characteristics that are divided into three dimensions: Values, Needs and the Big 5 (also known as the Five Factor Model (FFM))

1. Openness to experience

2. Conscientiousness

3. Extraversion

4. Agreeableness,

5. Emotional range

It’s built on the MEAN Stack — Mongo, Express, Angular and Node. It uses the O3 Labs HTML / CSS Front End Style Guide. All of this is open source on on O3’s Github.

It tells you a little about how we came up with our recommendation and ask you if we were right which helps train our Barista Bot.

Our next Lab was Moods. Moods merges skills and learning from 2 previous projects and brings them out of the web and into the real world. It uses personality insights gained again from twitter to change the mood of the room by changing the lighting color to something we think you’ll like based on your personality.

Moods uses:

* MEAN Stack

* O3 LABS UI Kit / Style Guide

* Philips HUE API enabled bulbs

We’ve taken Hue way further than just dimming like we did with Roombot and all the way to setting different color mood lighting.

So, what’s next? Well obviously more innovative and fun Labs!

But, we’re looking at this as more than one-off applications. One of the coolest things about Labs is that we get to explore new tools and technologies, with the potential to bring real value to our clients. What can start as a fun app to predict coffee preferences or set the mood can evolve into something that actually understands consumer needs and behaviors. And this can be applied to customer segmentation, product recommendations, delivering customer care, talent recruiting, financial risk profiling, and of course advertising where this is the current state of the art and soon to be standard. Future generations won’t even understand the current billboard style of just casting a wide net and hoping it gets to there right people and resonates. The standard will become targeted ads customized for user personality and preferences.

Thanks so much

--

--

--

Engineering Director @linode, @Awesome_Philly Board Member, @CTOSchoolPhilly Co-organizer, bike rider, beer brewer, coffee geek

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

How to Migrate from PipelineDeals to Salesforce in a Few Clicks [Tutorial] 2018

How to convert an HTML document file to text (TXT) in Java

Optical Character Recognition(OCR) — Image, Opencv, pytesseract and easyocr

iOS simple painting App — demonstration in Swift, SwiftUI and RxSwift

Simple painting App preview

Great work, Jovan! It's very useful to have an example code to start with ML on GCP!

Event Sourcing: What properties should Domain Events have?

Unity Game Development — Sound Effects in Unity

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Howie Ross

Howie Ross

Engineering Director @linode, @Awesome_Philly Board Member, @CTOSchoolPhilly Co-organizer, bike rider, beer brewer, coffee geek

More from Medium

‘Load na Dito’: Experiencing Data on the Internet in the Philippines Through Social VR

A screenshot from a webVR experience on the internet in the Philippines.

What makes a piece of tech disruptive?

Can a Non-Philosopher Read Kant?

How Artificial Intelligence Can Enhance Workplace Safety