The Entrepreneur: Ofri Ben Porat, Edgify

Edgify is striving to make AI more accessible, enabling any company from any industry to train complete deep learning and machine learning models, directly on their Edge devices.

About Us

Startups was founded over 20 years ago by a serial entrepreneur. Today, our expert team of writers, researchers, and editors work to provide our 4 million readers with useful tips and information, as well as running award-winning campaigns. We interview successful entrepreneurs from all over the UK, celebrating their achievements, hard work, and determination to get to where they are today. This article was authored by:

Founder: Ofri Ben Porat
Company: Edgify
Website: https://www.edgify.ai/

Edgify specialises in using “edge devices” such as MRI machines, supermarket checkouts, connected cars, and mobile devices – essentially anything with a CPU – to train deep learning models for a range of industries, negating the need for use of the Cloud.

Ofri Ben Porat, founder and CEO of Edgify, talks to Startups about business challenges, diversity in the workplace, and how he turned down one of the biggest New York-based VCs because of DocSend.

The Business

Describe your business model and what makes your business unique:

Edgify generates revenue by signing partnerships with retailers to roll out their technology across their network of SCO (self-checkout) machines. However, the company has plans to apply its computer vision solution across other industries, including healthcare and manufacturing.

Using an edge device for analysis and training of information, rather than the Cloud,  reduces the risks, costs and time associated with transferring sensitive-data to and from an external server. This allows businesses to train on the entirety of their data, and reach accuracy levels never achieved before.

Given the legacy technology at the core of most stores, Edgify identified retail as an industry that would benefit from its computer vision solution, particularly around the accuracy of self-checkout machines in grocery stores. Edgify’s solution can be implemented throughout supermarkets, in any scale or scanning device – and can even distinguish between similar barcodeless fresh groceries, such as Granny Smith versus Pink Lady apples. It then shares the acquired knowledge or model across a distributed yet collaborative framework of point of sale machines, leading to faster, more accurate and low-touch checkouts.

Check out the video below to see how Edgify works:

What is your greatest business achievement to date?

Surviving the first year post-pivot and managing to close another round after it – without any hate from the investors – is right up there. I’d been working on Pixoneye, a Computer Vision-based product with the ability to analyse personal photo galleries in your phone, for three years. We’d raised Series A and built a team of over 35 employees from the ground up working amid a driven, family atmosphere. Half – our commercial team – were in London. We’d started to gain traction. And yet in October 2018, I had to ‘fire’ them all, which was the hardest thing I’d done as a CEO.

My brain was fried. I’d been so hell-bent on growth mode, worrying about scaling and our revenue models and the realisation that this chapter was done hit me hard. The background to this was my co-founder and CTO Nadav Israel, who had come to me months before to say the research team in Israel had built an edge computing solution that was way more interesting than the product we were developing this technology for. The Eureka moment was the realisation that we could adapt what we’d developed for phones to run AI training on any hardware. The Board agreed with the direction.

I can discuss it now as Edgify has raised a significant seed round and proved the concept. We had some amazing people who we loved – and I’d work with again tomorrow – but we had no product to sell at the time. Communication and fairness were key and we reached out on their behalf to anyone we knew who was hiring, placing two of our team in portfolio companies of our investors.

How did you fund your business?

In October 2020, Edgify announced a $6.5m seed funding round, backed by Octopus Ventures, Mangrove Capital Partners and a semiconductor giant.

What numbers do you look at every day in your business?

In all honesty, at the moment, it’s probably the number of meetings in my calendar or the emails I need to churn through in my inbox. But more meaningfully, I’m in constant contact with the team and our top six clients, looking at the performance of our technology and the accuracy of our models.

I like to see how many new grocery items are added to the model each day for example – such as an unusual type of watermelon. Looking at these analytics also reveals where we’re managing to reduce theft in stores through false selection, which is exciting and reassures me that our technology is having the desired effect.

To what extent does your business trade internationally?

This is an enormous focus. Every major market has prominent supermarket or retail chains with self-checkout machines, so as a result, we’re already actively trading in the US, the Nordics, Baltics, Japan, France and Israel, as well as the UK.

Where would you like your business to be in five years?

If I’m still the CEO of Edgify in five years’ time then we probably have an issue. I expect to be relaxing on a beach in LA, sipping on an Old Fashioned. In fact, I’d like us to be the ‘vision’ technology division of a semiconductor giant – so we’d probably be going by the name of a name the world is far more familiar with.

If I’m still the CEO of Edgify in five years’ time then we probably have an issue. I expect to be relaxing on a beach in LA, sipping on an Old Fashioned.

What software or technology has made the biggest difference to your business?

I’d have to say DocSend had a pretty significant influence on the business and there’s a story behind that. If you’re not aware already, there’s nowhere to hide with DocSend. I turned down one of the biggest New York-based VCs because an associate there did not take our appointment seriously – despite the head of investment at our lead investor making the intro.

She tried to pull out of a meeting that had been in the diary for two months at the very last minute. It was only when I looked at DocSend I could see she had opened my email just two minutes before she emailed to say she wasn’t feeling well and would be heading home. I was able to see which slide she was on and the pages she’d spent most time on. 0.01 seconds on the proposition and 40 seconds on the revenue table – and that was the last thing she looked at. She didn’t look at why the revenue was where it was at that point, or what we were working on (see the pivot story above). It’s pointless just focusing on the revenue slide if you haven’t taken the time to understand the bigger picture.

Anyway, I ignored her email and showed up at their offices, where I found her to be looking like the picture of good health. She quickly gained interest and I was then introduced to two Partners of the fund. They offered due diligence and pre-terms, but ultimately I walked away.

Entirely separate to that, I’d also say Monday.com is essential for all our internals – but there’s less of a story there.


Growth Challenges

What is the biggest challenge you've faced in business?

Diversity – we’re still failing miserably and it’s a constant source of pain for me. Tech leaders convince themselves that the industry doesn’t have enough diverse talent, particularly when it comes to women in technical roles and I gave myself that ‘get out’ too. My co-founder is our CTO and I largely took his word for it, believing that it really was too hard to find and hire highly qualified and experienced female developers. Startups have to fight for talent as it is, so you find yourself succumbing to other pressures and not addressing the elephant in the room. As founders, we end up perpetuating the problem.

The reality is you just need to push harder to make it happen and now we’re learning the hard way by trying to undo the workplace culture and hiring processes we’ve had in place. I remain embarrassed about where we are, despite progress. At least with a professional HR manager we have now started to establish best practices to remove conscious and unconscious biases, while changing aspects of the hiring process that potentially alienate female candidates.

For example, one of our tests required candidates to deduce a program based on the raw code, which was often a game such as Mortal Kombat or Monopoly. Although this didn’t discriminate on the basis of gender in itself, it was an inherent barrier to people who were not interested in gaming – meaning recruits tended to end up being more like the ones we had already.

What was your biggest business mistake and what did you learn from it?

Not making MRR (monthly recurring revenue) my North Star. We chased growth and funding before we had product market fit pre-pivot. We also failed to vet some of our early investors properly. Someone once said to me that taking VC investment is like a Catholic wedding – there’s no divorcing. Chemistry really is everything as you effectively go to bed every night with your investors and need to know who will be on your Board when you enter discussions.

If they want to, VCs can hold you by your jugular and I went through that where one sitting Partner applied almost unbearable pressure. This can apply to resellers too – especially when they feel they can sell your product better than you (reality check: they can’t!). The joy of Edgify is that I’ve had the privilege of being able to fix all the mistakes I made with Pixoneye.

What one thing do you wish someone had told you when you started on your business journey?

The most important thing is the first paying client. Build a business on capital not equity. And in reference to the above, get a female co-founder.

How has the pandemic affected the market you operate in?

It’s accelerated it by forcing a lot of automation into everyday life. Our solution is centred around getting people out of stores faster, with reduced need to touch things, and generally better execution. The pandemic has demanded that of us all. Where it has complicated life for us has been our ability to do face-to-face integration. Ours is complicated software, so we’ve adapted to give demos to potential customers remotely – even setting up our own mini-market in the office, complete with scales and a pretend check-out machine. In the absence of being able to travel, we also packaged up our software to plug and play on prospects’ own hardware.


Personal Growth

Did you study business or learn on the job?

I never studied business. It was bartending school for me. I grew up on three continents, served six years as a Captain of infantry at the Israeli Defence Forces, studied conflict resolution and crisis diplomacy in Israel – and while doing so I launched a career in night-life, opening bars and gastro pubs. Tech came later when I started an online student services website before meeting Nadav. The cocktail training has come in handy. There’s an inestimable number of Negronis you need to be able to mix to save cash on lonely nights as you go through a funding round.

What would make you a better leader?

I’m a born leader. Ok, I’m joking really, but I genuinely think being ‘a leader of people’ is not the part that needs addressing – at least not first. Like anyone, I could do with a bit more knowledge. I’m the CEO of a deep tech company and am surrounded by geniuses with the deepest understanding of AI technology. I have enough to get by and am a communicator, but something like learning a programming language intimately would give me an extra edge. Otherwise, like every founder – more Zen, less stress.

One business app and one personal app you can’t do without?

Personally, I’m not sure I’d survive without Gett (previously known as Get Taxi). It’s the go-to app for Fortune 500 companies – and also me. And such is my need to put as much distance between the interminable talk of lockdowns, quarantines, red, green and amber countries, Booking.com is my mental health survival app of choice. I need a holiday and the app is incredible. You can book 50 hotels in 50 countries and cancel them all. But way above that is the level of customer service they offer, for which I’m a huge sucker. On a night I reached my hotel after midnight to find it locked, they chatted and offered to stay on the line for the night, while also somehow getting through to the hotel to get them to open up and give me a room to stay in. Going the extra mile doesn’t begin to cover it.

A business book or podcast that you think is great:

‘The Best Interface is No Interface' by Golden Krishna. It’s an ‘oldie’ now, but remains incredibly apposite, picking up on screen saturation and how our relationship with technology needs to change in an ‘app for that’ world. The podcast ‘Without Fail', hosted by Gimlet Media’s Alex Blumberg, chronicles super-interesting stories each episode from people from all walks of life sharing how their biggest bets went spectacularly wrong. It really puts things into perspective, and as entrepreneurs, we so often need to treat those two imposters triumph and disaster just the same, to paraphrase Kipling.

Ross has been writing for Startups since May 2021, specialising in sustainable business and telephone systems. He also runs the successful entrepreneur's section of the website.

He's previously written for Conde Nast Traveller and the NME and is passionate about music, sustainability, and travelling. Follow him on his Twitter - @startupsross for helpful business tips.

Leave a comment

Leave a reply

We value your comments but kindly requests all posts are on topic, constructive and respectful. Please review our commenting policy.

Back to Top