Select Page

Projects

A few projects from my archives

Here is a sample of my recent technical projects. I have left out the strategic framing projects that went alongside most of these.

The projects described here all involved the development of some novel core technology. These projects required sustained effort and investment. However, I have conducted numerous short-term vision or feasibility exercises for clients, such as Acision, Navteq (acquired by Nokia), Microsoft, Vodafone, Naspers, GSMA, Ericsson etc.

I am also a partner at Interpret Design (ID), a design-thinking business focussed on the design-led application of AI. If, prior to deploying AI or similar cognitive technologies, you first need to figure out your AI-contextualized product strategy using design thinking contact me.

AI Chat Bot – “Finance Speak”

My current project is designing an AI bot that can understand all things “consumer finance”, especially lending. In some areas, such as HELOCs, the average customer is unsure of the product and its suitability. Customer issues might be addressed via a conversation. Calling an agent is too much friction whereas chatting to a regular “support/faq bot” is too frustrating due to their typical lack of sophistication.

Much of my patent-pending work is trying to create a so-called Level-3/4 bot with advanced conversational capabilities. A lot of the novel work I have done is in coping with dialogs that stray from the happy path and involve having to deal with tracking the user’s attention as it meanders from subject to subject. Techniques involved include a novel application of embeddings in a conjoined vector space inspired by the Starspace algorithm. Additionally, a blend of supervised and unsupervised embeddings can significantly improve the handling of ambiguity in customer intent.

 

Blockchain: Retail Lineage

Combining my expertise in blockchain and my experience in e-commerce, I provided hands-on technical expertise to tackle fraud in the multi-billion luxury goods market.

There are two types of fraud – blatant knock-offs and deceitful trading of goods claiming them to be originals. The latter has led to specialized websites like Flight Club who act as an intermediary to vouch for authenticity.

I developed blockchain protocols to produce transferable point-of-sale digital receipts. But the project was framed by an ecosystem approach whereby consumers and merchants could trade in a secondary market.

ML: Dynamic Pricing

I consulted for the Credit Risk team in Prosper Inc. to explore the mathematical and computational foundations of elastic pricing for unsecured loans.

Pricing of loans by lenders is typically a function of credit risk, but seldom a function of demand and other elasticity factors.

My job was to interpret the foundations of pricing credit products using dynamic methods to the point that our credit team had a basis for a research project. Most of my contribution was in parsing the dense literature and interpreting some of the key mathematical principles into Python code.

Computer Vision: Financial Document Verification

I was asked by Prosper Inc. to look at a nearer-term problem of automatic document verification for online lending.

Using a novel ensemble of computer vision techniques, I built a patent-pending image-processing pipeline that, unlike conventional “check deposit” systems, could analyze financial documents at any angle and with any degree of background noise, such as a user might casually snap with their mobile camera (unconstrained). Under such conditions, my solution out-performed leading vendor solutions including Lead Tools and Google Cloud Vision. The end result is an easier and quicker application process for customers.

Blockchain: Lending IaaS

Prosper Inc. pioneered marketplace lending, connecting individual lenders with borrowers. Using digital orchestration and ML-based credit rating, unsecured loans are closed quicker than banks.

But what’s next? I developed a set of blockchain protocols that would allow frictionless instant lending, and even pre-emptive lending via “smart money” (read my article on Medium). This is more than just publishing loans on chain. That would be like storing photos of dollar bills and calling it “Bitcoin”. Hardly!

The project was part of an engagement to explore technological futures and IP generation as a future-proof vehicle for growth via category-defining user experiences (instant “smart” loans).

AI: Perceptual Color Harmony and Natural Language Generation (NLG)

Color sits at the center of any decor project. However, the computational, biological and perceptional reality of color are very different. For example, it is insufficient to merely count the number of pixels in an image and use the most frequent color to characterize the art, as is done by naive algorithms. Color perception is highly influenced by saliency, or what we find interesting in a picture. That said, what we find interesting has a perceptional and biological reality. It is possible to hang a picture in our lounge, glance at it a hundreds of times and yet fail to notice its true (veridical) colors.

Working at Art.com, I helped develop some of the world’s best color technology that used perceptional techniques to match colors. Additionally, I built a Natural Language Generation (NLG) algorithm that could describe the colors of any piece of art using the vernacular of an interior designer (using a language model derived from interior design blogs). 

Hardware/IoT: Digital Art

After extensive analysis of how to solve the Discovery problem in art, I helped create a skunkworks project to build a novel digital art display.

Our labs team create a hardware device, the embedded software and a mobile app at lightning speed. We also created a novel file player format (called .art) so that artists could make long-playing artworks that evolved over time (up to 1 year). And, to bootstrap the art platform, we created an extensive network of early adopter artists.

The pilot was a success with excess demand for devices and a number of key patents.

Design Innovation: E-commerce

Discovery of art and decor (or any aesthetic product) is complicated by the tendency of users to say “I know what I like when I see it“. As consulting Chief Scientist, I led comprehensive design research programs using AI, neuroscience, neurobiology, psychology and ethnography to find foundational (and computational) methods that could predict what a user might like upon seeing it. [Image shows tastes in art as discovered via social curation networks in the data.]

The work was done in a “innovation lab” that I directed. Some of the concepts are documented in this slide share.

Among other things, I invented a method to discover the theoretical limits of how quickly a user could “like it when they see it”. This was an extension of X’s work at MIT.

The work generated significant IP for the client that significantly impacted valuation multiples at the time.

IoT/ML/AI: Sensor Networks

McLaren is a tour de force in the world of high-precision F1 racing and luxury sports cars. McLaren’s brand is synonymous with “performance engineering”.

I was hired by McLaren’s Applied Technologies division to extend the reach of “performance engineering” to new clients (and revenue streams) by architect a sensing platform that could ingest data from any number of IoT devices and process the data in real-time using ML, AI or functional programming.

The work exploited McLaren’s reputation with real-time sensing of F1 racing cars. The goal was to  attract clients in the “performance sports” industry (e.g. Nike, Specialized etc.)

The platform and vision were central to the initial growth and reputation of the MAT division.

Labs: Building at the “speed of thought”

I the technical co-founder of an “intrapreneurial” unit inside of O2 (UK) and Telefonica R&D (Spain) where I formulated and executed the idea of “Operator 2.0” — the translation of Web 2.0 methods and tech stacks to a carrier business. Projects I created include:

#Blue – a text-messaging API built using “lean” methods and modern software techniques at 1/10th the cost of historical carrier product costs: one of the earliest applications of “Lean” inside of a large org (use case shared with Eric Ries.)

O2 Incubator – the first start-up incubator created by a telco in Europe, sponsored by Sir Simon Devonshire who went on to run Telefonica’s Wayra London.

connFu — see next project.

During this engagement, I evangelized the concept of Connected Services and wrote a book to explain it (read review here). It has been widely read by carrier executives globally.

Agile: connFu

Whilst consulting for O2 UK, I approached the R&D division of Telefonica to propose a radical new approach to building carrier services: connFu.

This was both a platform and a scripting language (“Domain Specific Language” – DSL) — see example in adjacent image.

connFu enabled developers (internal/external) to create telco services “at the speed of thought” – i.e. as fast as someone could code, versus the old carrier waterfall method that hindered innovation.

The telephony services were powered by a telephony server product (Voxeo) atop of the Jajah IP-telephony network (strategically acquired by Telefonica).

This was the first – and seemingly only – time that a carrier built its own programming language and fully embraced the APIs and modern software techniques that power the web. The exact same approach is how Twilio built a billion dollar company. (Note: I had recommended to Telefonica that they acquire Twilio and made the necessary introductions, but it wasn’t to be.)

 

Labs: Mobile Mashing Lab

Motorola hired me as a technical “co-intrapreneur” to help build a brand new business unit that offered scalable apps services to carriers. I assumed the role of Chief Architect.

Besides providing the entire apps strategy, I was hands-on with architecting numerous solutions, including content management systems, mobile TV platforms, voice services, push-to-talk and music streaming using SIP signaling extensions.

I also created a “Mashing Lab” in Chicago for building service mash-ups. We built the world’s first location-based taxi-hailing system that also included single-click voice connections between driver and passenger. It was Uber before Uber.

Subscribe to My Newsletter

An occasional digest of my posts and interesting things at the intersection of AI and business. Absolutely no bullshit and complete privacy.

 

Subscribe ToMy SimplexityNewsletter

Join our mailing list to receive the latest news and updates from my blog and projects.

You have Successfully Subscribed!