stub Maria Elena, Director of Solutions at Stradigi AI - Interview Series - Unite.AI
Connect with us


Maria Elena, Director of Solutions at Stradigi AI – Interview Series

Updated on

Maria Elena Carbajal brings over 25 years of Artificial Intelligence, Information Technology and Telecommunications experience to her professional career. She has 18 years of experience working at a telecom company in Canada and Sweden, as well as in the Aerospace, Energy and Information Technology areas with various companies. Currently working for Stradigi AI, an Artificial Intelligence leader in Canada.

Maria Elena Carbajal has performed in many functional roles within R&D, Engineering, Global Professional Services, Digital Transformation and Information Technology. Her international exposure includes working and managing teams in countries such as Peru, Canada, USA, Mexico, Brazil, Sweden, Finland, Norway, Russia, Estonia, and Belarus.

What is it that attracted you to AI initially?

I have always been passionate about working in the tech sector. As an individual, I’m always invested in optimizing what’s around me: from organizing my household to bringing efficiency to my teams, clients and businesses in general. It’s at the core of my being. I’ve been fortunate to have very hands-on experience with implementing leading technology in the past two decades in various industries, so skills become highly diverse and transferable.

Looking more specifically at AI, I’m a strong believer that Artificial Intelligence and Quantum computing will revolutionize all industries — no exceptions. AI is critical to unleashing and fueling optimization in all areas: business, professional, and personal. That’s what drew me in and kept me engaged and inspired, day after day.

You were previously employed with Ericsson for 18 years, what made you decide to join Stradigi AI?

I was interested in focusing my professional efforts on AI due to how much it will impact and change the way we live and work. AI opens the door to a range of problems you look to resolve when working with enterprises big or small that gives you the chance to make a difference, move the needle, and use technology for good. Plus, I had extensive executive experience at Ericsson, which, by contrast, is a massive organization. Transitioning to a startup like Stradigi AI, I wanted to push myself out of my comfort zone and experience what it’s like first-hand to be part of the thriving, evolving AI ecosystem that’s forming in Montreal. There’s something motivating and energizing about being a part of this community.

I’ve been at Stradigi AI for a year now, and after a year of working with an amazing group of professionals and innovators, it’s clear that my experiences at Ericsson have been so valuable and transferable to any organization, no matter the size. My philosophy was always about moving the technology needle one client at a time, and I’ve brought that to Stradigi AI, too.

Stradigi AI allows someone with no machine learning experience to produce AI models, can you describe how this is achieved?

A lot of the discourse in the AI and ML world is around “democratization.” Which, to put it loosely, is all about making AI available to the masses. But availability and usability are not the same. With our self-service ML platform, Kepler, our primary goal is to ensure that internal SMEs and analysts can produce models with advanced ML techniques, without having to learn complex data science work, or involve their data science teams.

From a practical and technical perspective, this is achieved through automating the step-by-step data science processes that typically take time and expertise to complete. For example, Kepler automates the feature engineering process, a complex, multi-step endeavor. It also automatically creates a pipeline by selecting the best algorithms, undergoes automated configuration and hyperparameter optimization — all hands-free.

The goal of having this level of automation in the process is to free your experts of wasting time on trivial and time-consuming tasks. By automating these steps, Kepler gives your team more time to think about the next grand innovation, rather than the menial tasks of the day. For analysts and SMEs, it’s also a route to upskilling: by implementing ML tools into their day-to-day, you’re giving them the chance to enrich their analysis and therefore their approach to key use cases.

What are some interesting AI models that you have seen from companies using Kepler? 

The beauty of Kepler is that it covers a wide range of use cases across several industries, using an array of techniques from classical ML to deep learning. From governments to investments, Kepler can help leaders achieve measurable results.

A few impactful projects that come to mind that have a large impact on the way we live and work, is the development of innovative models in the healthcare sector, where we used image segmentation models and regression models to detect illness. Another is our work with regulatory bodies in local and national governments in using Natural Language Understanding to categorize complex text information, and bring new efficiencies to legacy processes.

On the other end of the spectrum, we’ve also leveraged Kepler to optimize trading activities for a client in the financial sector.

Stradigi AI uses an Automated Data Science Workflow. For those who are unfamiliar with this, can you describe what it is and how it’s used by Stradigi AI?

Automated Data Science Workflows (ADSW) are the end-to-end data science processes that work within Kepler. ADSWs were created to solve a number of use cases, so we built each “workflow” to have highly practical applications. For example, one of our workflows is Time Series Forecasting, which allows professionals in CPG or retail to predict when inventory will need to be restocked. There are eight pre-existing workflows in Kepler, which are all intuitively designed for the non-data scientists.

ADSWs are advanced ML workflows that automate key processes, some of which I alluded to above. In an ADSW, Kepler automates:

  • Hyper parameter optimization
  • Configuration
  • Model selection
  • Training and testing data splits
  • Dashboard creation
  • Evaluation of model metrics

All the user needs to get a deployable machine learning model is data and a use case to tackle. And, depending on the data set, all of the complex work within an ADSW can be completed within minutes.

What types of data can be used?

The Kepler platform lets you work with tabular, text, and image data.

For those that aren’t familiar with data and data types, I’ll break them down:

  1. Tabular: this would be a spreadsheet containing key information like sales data, or a database table of client demographics, products, etc.
  2. Text: this type of data can come in so many forms, think emails, customer reviews, social media content, library archives, contracts, etc.
  3. Images: think of product galleries, or photos of items on an assembly line.

Video data will be ingestible in Kepler in the future. On our website under “Data Types” we explain what types of data can solve key use cases. You’d be surprised how much data goes unleveraged, especially in larger enterprises.

Do you have any tips or strategies for women who are interested in joining tech?

I have three tips that I think are truly fundamental to success for anyone to thrive in the tech industry.

1 – Learning. This should always be a part of your life. No matter how young or how old you are, you should always have something to learn. It doesn’t matter how you go about learning or who you’re learning from, just be ready to receive knowledge. Open your mind. Clear your brain so you can be ready to receive more knowledge, more love, more empathy… just more. Be obsessed with your own development. A great reminder is that being ready to learn is one of the key manifestations of empathy.

2 – Passion. Examples of hard work have never fallen short in my professional experience. I’m always ready to raise my hand at work to take on complex situations or complicated activity. The more I do that, the more I realize that everything is possible. I wouldn’t hesitate to jump out of my comfort zone and take on that extra challenge. When you do this, leaders can identify and appreciate the passion you exhibit.

When you approach work this way, you do not need to wait for that great job opportunity that will change your life. If you pay attention, you will notice there are plenty of small tasks around you that will give you more exposure to decision makers, and most importantly, more possibilities to learn.

3 – Mentorship. For me, mentoring is such a powerful tool, as it flexes your listening and learning muscles or skills. Mentorship can also bring you closer to great leaders from your professional networks or within your personal circle. Throughout your career, it’s crucial to identify the kind of leaders you can trust and follow, and select them as your mentors and role models.

The leaders that believe in you will push you out of your comfort zone and be there to help you gain strength, too. Great leaders and great mentors can be brutally honest, but they can also be excellent listeners. Finding selfless people to help you reach your full potential can offer you some of the best teaching moments of your life. Now, your task is to find and recognize who these mentors are or could be for you — and trust them.

Thank you for the interview. Your three strategies for those wishing to enter tech are applicable to anyone and I completely agree with them. Anyone who wishes to learn more about Kepler or this amazing company should visit Stradigi AI.

A founding partner of unite.AI & a member of the Forbes Technology Council, Antoine is a futurist who is passionate about the future of AI & robotics.

He is also the Founder of, a website that focuses on investing in disruptive technology.