Startup Genesis: US vs. Global Perspectives with Ned Lomigora

In the ever-evolving landscape of entrepreneurship, success often seems like an elusive prize, attainable only to the fortunate few. But is it solely a matter of chance, or is there more to the equation? In this exclusive interview, we delve into the remarkable journey with Nedžad Lomigora, CEO, CapeAnn Tech Director, MIT Enterprise Forum NYC, USA, whose story embodies the confluence of good luck, seizing life's opportunities, and even a special connection with NASA. Join us as we explore insights on venture capital investments in the Balkans, the distinctions between startup formation in the United States and the rest of the world, and the future of businesses in the age of rapid AI advancement with a new keynote speaker at TechHosted09.

Considering your journey and development from being an athlete in Bosnia and Herzegovina, an Olympian in the 1994 Winter Olympics, to one of the few foreign nationals who managed to study at MIT, how do you choose the path to success, and what is the required mindset and approach to life and opportunities?

Ned: I’ve seen many try to explain their path to success by invoking hard work and sacrifice. If they were honest, mo st of them would simply admit that luck is the most important factor when it comes to success. My luck was that I was training luge well before the war in Bosnia started, which secured me a position on our Bosnian Olympic team that was sent to the 1994 Winter Olympics in Lillehammer. That enabled me to leave Sarajevo during the war, and then after the Olympics, I was invited by the US Luge Association to come to the US. Once in the US, I met several people who played a key role in helping me get into a US college, and only then was my hard work and dedication to excellence responsible for a full scholarship to MIT.

The mindset that I had was that it couldn’t get any worse than being in war, that no matter what challenge I was facing, I wasn’t intimidated, and I was hungrier to succeed than anyone around me, even though MIT is the most competitive school in the world.

You began your career during your studies, when you were “challenged” by close individuals to create and contemplate ideas that could solve tasks for those who knew how to use technology to provide acceptable answers.

What was the first software you worked on like, and when did you realize its true potential and purpose?

Ned: My first original contribution to technology and a breakthrough came from finding a unique approach to solving a problem. I never set out to discover or develop what I ultimately developed. I was working on a tele-robotics-related project at the MIT AI Lab, trying to figure out how to control a remote robot (it was simulated via a real robot I built that was in a different lab at MIT from where I was controlling it). The challenge was to use feeds from multiple cameras placed in a room with the robot to triangulate distances and coordinate both the robot relative to various objects and the objects relative to the boundaries of the room itself. Based on those distances and relative positioning, the remote robot operator would then issue specific commands to the robot using coordinates derived from an algorithm that used those coordinates.

This approach was state-of-the art at the time but was very costly due to the several cameras that were required, the raw processing power (expensive hardware), and the knowledge of specialized fields like machine vision (which took years to master). I wanted to find a solution that would be commercially viable (at a reasonable cost) and didn’t want to spend years mastering machine vision. My original insight was to flip the problem upside down: instead of using cameras to track the robot and objects in the room, I came up with the idea to tag the robot and all the objects in the room and track their relative positions via a field of wireless readers that read those tags’ locations and triangulate their unique positions relative to each other. I also decided to use RFID as the basis for my approach.

This turned out to be a true breakthrough because it was several orders of magnitude cheaper to implement and did not require specialized knowledge of machine vision because there was no need for cameras or expensive equipment. RFID technology at the time was not used for those purposes, and it was not very versatile, which, over the years, produced a slew of companies building RFID tags and readers. After that, the technology found a completely new use to track various objects in logistics, inventory, manufacturing, supply chain, distribution, and many other applications.

NASA is the most prestigious agency in the US (probably the world) and also the dream of many engineers. You have a chance to become a part of it; can you tell us why you turned down an offer you once received from NASA?

Ned: NASA's Jet Propulsion Laboratory (JPL) in Pasadena, CA, extended me an offer to work there on a special project they said would perfectly fit my skill set. At the time, they could not reveal what the project was but convinced me that the work on building a tele-robotic interface I was doing at MIT was exactly what they were looking for. I built a virtual reality interface that could control a remote robot navigated by using RFID technology, considering a time delay in a feedback loop between the operator and the robot.

The position would entail a lead role on the said project. NASA is the most prestigious agency in the US (probably the world), and as an engineer, I was honored and humbled by the prospect of working there. However, at the same time, I was offered a position at multiple tech and Wall Street companies. My goal was to work at a company where I could learn as much about running a business as possible while still being connected to technology. While NASA is the most challenging and intellectually rewarding place to do engineering at a massive scale, it was not the place for me to learn how to build or run a business, which drove my decision to take a role with a tech company that was a perfect fit for me.

Interesting fact: the secret project I was supposed to lead at NASA turns out to be the Mars Rover Project, and I was to build the virtual reality interface for it using technology I built at MIT because it would be perfect for a time-delayed feedback loop between Mars and Earth (It generally takes about 5 to 20 minutes for a radio signal to travel the distance between Mars and Earth, depending on planet positions).

From your personal experience, what does it take to build a tech startup according to US standards?

Ned: To build a tech startup according to US standards, the founders have to have a good understanding of how venture investment (either venture capital or angel) works, who their target market and customers are, how to build a minimum viable product, how to find a product market fit, how to hire the right employees, and how to reward them properly. These are some of the tenants of any successful startup in the US.

You have provided global software consulting for clients on the Fortune 100 list. Can you give us some insight into what this entails?

Ned: Over the last 20 years, software outsourcing has grown in the Balkans to be the dominant sector in the IT industry. Most of what is done is a fee-for-service model that distances outsourcing companies from the end client and focuses their work only on executing products and services already designed and architected by others. This model is based on pure arbitrage of labor: the local cost of living in the Balkans dictates local salaries for IT professionals to be significantly lower than salaries in West Europe or the US. Outsourcing companies in the Balkans compete primarily on price by offering their services at a cost that is lower than the cost of equivalent services in the US and West Europe, making it enticing for clients there to outsource to the Balkans.

The problem with this model is a lack of access to end clients, which means a lack of opportunity to be exposed to learning skills such as conducting market research, developing product market fit, developing a minimum viable product, and other essential skills learned only when working directly with the end clients. Our engineers are great at building what is given to them but not coming up with new product ideas, as they don’t have access to the US or West European markets and have never had the opportunity to build that skill set.

That is why I started 387labs.ai, a full-stack AI company that builds products instead of purely outsourcing skill sets. The future for Balkan IT companies is in building their own products and targeting global, not local, markets.

Big names like Google and Amazon invest large sums in technology, and in total, AI startups received $52.1 billion in venture capital in 2022. Given your interactions with experts from the Balkans through various organizations, what are your estimates and predictions for venture capital investments in the Balkans?

Ned: Combined VC funding for SEE startups has grown from $218 million in 2017 to $1.3 billion in 2022 (Q1–Q3 survey), a 6x increase. Startups founded and headquartered in Southeastern Europe are worth a combined $31.1 billion as of 2022. Enterprise Software, fintech, and Transportation are the leading startup sectors in SEE. The three industries combined have generated over $24 billion in enterprise value, or over 50% of the total value of the region. Within SEE, Romania and Greece are the leading countries in value creation. But Croatia has been the fastest riser in the last five years. The newer generation of startups is scaling faster than ever. The availability of local capital is growing. The pieces are in place for the next generation of startups in Southeastern Europe to build a world-beater in the Balkans.

You have advised numerous startup founders throughout your career. Can you tell us how tech startups are formed today, how they evolve, and what challenges they face?

Ned: There is a big difference between startup formation in the US and the rest of the world. In the US, there is an extremely developed infrastructure for startups that makes it much easier than anywhere else in the world to start a company and receive funding (except for the Arab Emirates and Dubai, where there is currently the highest rate of funded startups, especially US-based startups). Startups in the US typically form to address a challenging and massively profitable customer problem with the goal of doing it in a way that is superior (cheaper, faster, or better) to current solutions in the marketplace. Startups spend their initial time doing market research and discovering the product's market fit before they decide what the market wants. Technology comes second once the product-market fit has been established. Throughout their journey, US startups have access to various mentors, advisors, alumni networks (especially those from top-tier colleges), angel funding, venture capital funding, government grants, accelerators, incubators, and a slew of other resources that help them succeed.
Startups anywhere else have a much harder time compared to US startups. For example, in the Balkans, startups do not have a strong investment network, and those few investors in the region often lack technology, operational, and startup experience themselves, which makes it less desirable for startups to take investments from them. The reason that Falcon Venture Partners was able to establish itself as a recognizable VC in a relatively short period of time is because we are hands-on operators with deep knowledge of the US and West European markets, deep technology expertise (especially in AI), operational experience, and a remarkable track record of successful investments.

AI is rapidly transforming the startup landscape, and it's clear that its impact will only continue to grow in the coming years. What will the future look like for AI startups?

Ned: Businesses worldwide will be transforming their business practices to embrace and integrate AI. Because AI has become a must-have, just like the Internet was back in the 1990’s, many future startups will position themselves as AI-first companies. That means that AI startups as a class will not exist for too long, as most startups will be AI-first startups.

The bigger question is what business model AI startups will develop to be able to compete in the increasingly automated world of software and technology development. In the past, the critical factor for a company's competitiveness was their ability to attract the most engineering talent, mostly software engineering talent. In the future, AI will produce large amounts of software, so having a large software engineering team will not be as big of a competitive advantage as it used to be. On the contrary, any startup with a large software engineering team will raise questions among investors and customers about why the company is not using AI to develop software. This will open new ways of valuing the company and its ability to produce solutions. It will favor people with imagination and talents that are not engineering-oriented a lot more than pure engineering expertise. Those who can design great UI, those who understand how to design a product, and those who understand how customer behavior drives their habits will be much more valuable to AI startups than those who can develop great code. The future success of AI startups will paradoxically depend on their ability to not rely heavily on engineering expertise but on their core business skills and their ability to offload large portions of their technical responsibilities to AI.

AI is not the be-all and end-all, and it should be carefully applied when considering its use because, even from a purely technological perspective, AI may be more of a hindrance than a benefit when applied incorrectly. Many companies are hesitant to deploy AI out of concern that they don’t know enough about it to know how to make it work for them, and they are not sure what their ROI would be as it is such a new area of technology.