Sign In

Ivan Litovsky

VP Product | AI Engineer | 2x in $1Bn+ start-ups | Entrepreneur

Ivan Litovsky is a highly accomplished professional with a diverse educational background and extensive experience in various roles within the tech and product management industry.

He holds a Master's degree in Digital Business Strategy from HEC Paris and has studied Economics, Strategy, Marketing, Finance, and Information Systems at Télécom Paris. Ivan also has a background in Mathématiques Physique from Lycée du parc.

His career journey includes notable positions such as Former VP Product, Director of Product, and Head of Product at Ankorstore. Ivan has also served as a Product Advisor & Mentor, Building AI Products, and Entrepreneur In Residence in different organizations.

Prior to his roles at Ankorstore, Ivan Litovsky was actively involved in product management at Criteo, where he held positions such as Senior Product Manager, Analytics & Data Science Team Lead, and Data Analyst. He has also gained experience in finance and strategic analysis through roles at Clipperton Finance and internships at HelloFresh and Capgemini Consulting.

Moreover, Ivan Litovsky co-founded Ask Blink and contributed to the innovation and development of AI products at Stealth. His well-rounded expertise and hands-on experience in various facets of product development and management make him a valuable asset in the tech industry.

Highlights

Jan 4 · linkedin

AI got 2x more coverage in newspapers vs. 2022, and 100x more coverage vs. 10 years ago…  Yet, there's widespread misunderstanding about AI's impact, what it means for jobs, and where opportunities lie.

The confusion comes from the nature of AI:

1️⃣ It is technically very complex, so most people misunderstand how it works and make wrong assumptions about what it could do.

2️⃣ The term “artificial intelligence” is misleading. When most people hear “AI” they think about a machine that can think like a human, like in dystopies such as Terminator or Matrix. This further leads to poor assumptions about what it could do.

3️⃣ People assume it’s new, while AI has been around us for decades: It's been integrated into social media, e-commerce, advertising, manufacturing, transport, etc. and is utilised by >50% of the world's companies

4️⃣ People are fooled by the countless impressive demos posted on social (“a website built with chatGPT”, “edit photos with just prompts”, “take a photo of your fridge and an AI gives you what meal to cook”, etc.). These demos are way easier to build than people think, but they are just demos. It is much (much) harder to build game-changing applications that can be sold and can scale.

5️⃣ People assume “AI” is just one thing, while it’s thousands of different models, the majority of them open source, that are used in a wide variety of AI applications.

6️⃣ Actually, the math is simple and dates from around 70 years ago, but the implementation at scale is extremely complex.

7️⃣ There's a misconception that AI development is solely about building models. However, the true challenge lies in creating user-centric applications that effectively incorporate these models.

-- As I looked deeper into ChatGPT and Large Language Models (LLMs), I realised the complexity of truly understanding AI opportunities. Most of the existing content is either great but too technical, or oversimplified and wrong. I couldn’t find good content to help executives, entrepreneurs, or professionals ponder AI's impact on businesses, startup ideas, or careers.

So I decided to write more actionable content myself!

Stay tuned! I will post on Linkedin (short version) and on SubStack (longer version - link in comment to receive articles by email!)

-- About the data: # of articles published on their website I picked these 2 newspapers because they were the easiest to scrap. From what I've seen with manual verifications on other websites, it looks representative.

Nov 23 · linkedin

Underperforming new hires cost companies millions of €. But why did these new hires still pass the interview processes?

All the studies I found on the topic say the same thing: mis-hires are caused by a variety of reasons, equally due to mis-assessed hard skills, and soft skills.

It is aligned with my personal experience (hiring >50 people at Ankorstore), so I believe these studies are correct. Let's deep-dive:

-- PROBLEM:

1/ There is a difference between demonstrating hard skills during case studies (approximately a 1-hour long exercise) and in real-life (iterating for days, weeks, months on a project). => Candidates who speak confidently with decent technical knowledge can very easily impress during interviews. But in real-life, communication skills are much less important than being a doer, solving your own problems, being autonomous & proactive, going fast. These are hard to test in interviews.

2/ Soft skills are hard to assess in interviews. We can assess presenting skills, oral communication skills. But what about teamwork? Motivation? Attitude? etc. => I don't know how to do that during interviews.

-- SOLUTION:

An easy solution seems to be reference calls. From. my experience, >50% of candidates get rejected after ref calls... so doing more ref calls can reduce the mis-hire rate by ±50%. that's huge.

1/ Hard skills: former managers/colleagues can easily tell you the actual contribution to projects of candidates. 2/ Soft skills: former managers/colleagues can easily tell you if candidates are team players or had a negative attitude.

-- so why don't we do more reference calls? (For example, I managed 50+ PMs, Designers, Data, and I have been called for exactly ZERO references... sounds weird)

Ivan Litovsky
Add to my network

Location

Greater Paris Metropolitan Region