AI is here, and is here to stay. It’s revolutionizing everything. Industries are being transformed, and our way of life is evolving faster than ever. The future is unfolding right before our eyes, and those who leverage its capabilities will have a competitive advantage, while those who fail to adapt may struggle to keep up.

At the forefront of this revolution are three megaminds: Shedrack Mwiine, Timothy Ndung’u and Akash Dadhley. Each a genius in their own right, they represent the best of what the Strathmore Institute of Mathematical Science (SIMS) has to offer, bringing together the power of Actuarial Science and Mathematics, Financial Engineering, and Financial Economics. And as fourth-year finalists on the cusp of graduation, these three are already making waves, carving out their place in a new era of innovation.

So what is their mark?

Last year, they stepped up to the challenge at the Nairobi Inter-University Hackathon, an event that brought together some of the brightest minds from across the country, all competing under the theme: “Harnessing AI for Real-World Applications.” Despite coming from courses where programming isn’t the primary focus, they were determined to push boundaries and challenge the norm. For Akash and Shedrack, it was a chance to put their skills to the test and make a real impact: “It was an opportunity to apply what we know in a practical, impactful way and put ourselves on the map.” Timothy saw it as a perfect opportunity to prove himself and compete against the best in Nairobi. With their mission clear and their eyes set on the prize, they signed up and decided to compete in the AI in Management and Governance category, focusing specifically on finance. It was the ideal intersection where their mathematical expertise and passion for artificial intelligence could converge to create real-world, game-changing solutions. 

The Innovation: AI-Driven Portfolio Optimization

They built an AI-powered portfolio optimization tool, a sophisticated application designed to revolutionize investment decision-making. At its core, the app leveraged neural networks through the H2O package in R programming, enhancing the traditional Markowitz Mean-Variance Optimization (MVO) model. Unlike traditional MVO, which assumes fixed historical correlations and static risk-return profiles, the AI-driven optimizer introduced adaptive learning, real-time adjustments, and improved predictive power. 

How they developed their solution, Akash says “My background in financial economics enabled me to leverage a solution for the hackathon by building an AI-powered app that assists individuals in efficiently allocating their investment portfolios. This expertise provided me with a deep understanding of financial markets, risk management, and data analysis, allowing me to develop sophisticated algorithms and features that optimize investment strategies and enhance user experience.” 

With a solid foundation in financial engineering, Timothy developed algorithms that balance risk and return, ensuring the tool delivered optimal investment strategies, which were crucial in designing their portfolio. On his part, Mwiine says, “Actuarial Science, which laid a quantitative and financial foundation, along with my background in machine learning and artificial intelligence, allowed me to merge the three disciplines and translate complex financial concepts into exceptional code and a high-performance application.”

The app provided three key optimization models:

1. AI-Driven Portfolio Optimizer – Used neural networks to dynamically optimize portfolio allocations based on evolving market conditions.

2. Traditional Mean-Variance Optimization (MVO) – Maintained the classical Markowitz framework for risk-return balancing.

3. Weighted Model Selection – Allowed investors to assign confidence weights to each optimization method, tailoring their strategy based on their trust in AI versus traditional finance models.

Additionally, the app-enabled investors to select stock, bond, and commodity tickers, generating optimized weight distributions for each selected asset within its respective asset class. This feature made it a robust tool for both institutional and retail investors looking for smarter, data-driven investment strategies.

Juggling a high-stakes project with the demands of life, academics, and the whirlwind of fourth-year activities was no small feat. Building an application of this magnitude required not only technical expertise but also an immense amount of dedication and time. They knew the stakes, and despite the steep learning curve, their commitment to innovation kept them focused. They sacrificed comfort, pushed their boundaries, and remained driven by a shared vision: to create something that could change the landscape of financial governance. Late nights, endless lines of code, and the relentless pursuit of excellence became their norm, but in the end, it was all worth it.

Their hard work paid off. They emerged as the winners of the AI in Management and Governance category and were crowned first runners-up across all categories in the competition. But for them, the win was about more than just the title. It proved that innovation knows no boundaries and that with passion, resilience, and a willingness to step outside their comfort zones, anything is possible.

“What’s the harm in trying?” says Mwiine. “That’s always been my motto, and it’s worked for me.” For students hesitant to compete outside their field, the trio offers a simple yet powerful piece of advice: “Embrace the opportunity, step out of your comfort zone, and dive into new challenges. You never know how much you’ll grow, and you might just surprise yourself by succeeding in ways you never imagined.”

Congratulations to these three mega minds! #KeepWinning and #GoPlacesWithSU

Article written by: Jemmy M. Kamau