AI Student Ambassador Ngesa Marvin: A Community Advocate in Kenya

Predicting the exact location of harmful weeds in Lake Victoria is a challenge but with Google Earth and wind data, Intel® Student Ambassador Ngesa Marvin thinks he can help eradicate them. Ngesa started his journey into technology with a single HTML coding book and library of inspirational history on Edison, Tesla and Faraday. He is now using his education, skill and passion to combine Artificial Intelligence, Machine Learning and Internet of Things into electronic design and innovative prototyping that help his community in Sub Saharan Africa.

Tell us about your background.

I am passionate about driving positive change in Africa through Technology, Entrepreneurship and Innovation. I serve the Intel® community in my region as a Student Ambassador and the Google Developer Groups, in the Sub Saharan Africa, as the Internet of Things (IoT) champion where I am responsible for presentations, training and sharing contents on IoT, Machine Learning & Artificial Intelligence, and Electronic Design & Prototyping. I frequently speak at conferences, and I actively support developers, sharing my passion and experience by organizing events, publishing videos and tutorials, writing code samples and mentoring developers and startups.

I have a great desire to help creative people make great products that help people live a better life through technology, and building communities around technology ,so that people can learn and help each other to become better makers of great products.

I have always had a great yearning to design intelligent electronics devices. In 2015 I led a team to build a motion algorithm that interprets hand gestures to read sign language and convey messages in text and audio, allowing the hearing-impaired to communicate more effectively. The project won the top award at the Lions of Science 2015 Competition.

I have also worked for early-stage startups and started some on my own. Although they are not always successful, I relish the learning opportunities that come from my failures. In January 2015, I founded a startup company that aims to change the way people vote and make decisions and I serve as its CEO. Under my term, we have completed the first phase which includes software and the design-build of the hardware. I also led the startup to the finals of Pivot East Competition, East Africa’s premier mobile startups pitching competition.

I enjoy working with Python and JavaScript to control electronics and microcontrollers. Recently, I began to dive deeper into Machine Learning and Artificial Intelligence (AI), Android Things, Analog Circuits and Embedded TensorFlow. In School, I am working on a range of units including AI, Embedded Systems, Control Systems, signal processing, wireless networks and electronics. I am working very hard to become an expert in my field and change as many lives as possible.

I worked as a Radio Frequency Engineering Intern at Communications Authority of Kenya where I participated in onsite inspection of radio installations to ensure compliance with the license terms and conditions, analysis and resolution of cases of harmful interference to radio spectrum, radio frequency monitoring and surveillance, site engineering considerations and predictions of radio wave propagation in various parts of the spectrum. Earlier this year, I contributed to the Radio Network Planning, Optimization and Access efforts in Nairobi at Huawei Technologies (Kenya) Co., Ltd. I also shared my opinions and reviews on 5G, and the exciting inventions and breakthroughs to the development of 5G industry in my publications.

I strongly believe in leaving all that I interact with for the better of my involvement, in the potential of getting things done, in sharing ideas and in inspiring my team to do things which others once thought was impossible.

What got you started in technology?

I have been a maker since I was a young boy. In the village where I was born, there are a lot of tractors and in primary school, during break time, we collected clay and would build tractors and factories and play with them. It was our version of Legos*. When my family later moved to a different town, I was amazed by the cars my friends built using aluminum wires and the rubber from the inner tube for bicycle tires. I tried to build some on my own. I remember looking everywhere for wires and actually got electrocuted when I tried to pull electric wire cables from the meter box to make a car. I almost lost my life — it was a definite learning experience. By class five, I was busy building different dipole and Yagi Antennas for my small transistor radio to improve my reception and by class six I was drawing designs for a quadcopter.

When I joined the Nairobi School in 2009, I got access to a library with a variety of books and encyclopedias where I read more about Isaac Newton, Alessandro Volta, Nikola Tesla, and Michael Faraday and got extremely inspired by their work. At that point, I knew very clearly that I wanted to be an Electrical Engineer. When I got back to class, I shared everything I learnt with my classmates. As a result of this, I was selected to be a Physics Promoter where I was responsible for sharing Physics concepts and tackling exam questions in front of my classmates. I then formed a Physics Club within the school. In this Club, I helped organize students at different levels to teach others and at form four in 2012, the club organized the school’s first National Physics Contest. I spoke at the assembly to release Physics Hour results and advocate for my club activates. The lessons I learnt from high school played a very crucial part of where I am today.

My desire to program started 2011 when one day, at my uncle’s place, I accidentally got hold of a book which had some HTML codes inside. I tried to play with it a little bit following the instructions and I was amazed with the outcome.

When I joined the university, I found that it was a good idea to strike a balance between studies, interests and other things the university and companies had to offer. I formed a small team of focused individuals inspired to design, build and accomplish simple projects like blinking LEDs to others with far reaching impacts. Over time, we have become obsessed with what we do.

At my second year of study, I applied for Intel® Student Partner program and was privileged to be chosen. As a student partner, I enjoyed sharing my zeal for Intel® Architecture, engineering and open source electronics with fellow students. One of the most important parts of my work is when a developer or an engineer who has never worked on Intel® architecture before learns how to get started. Passion, mentors, companies, the team and my community have kept me on the path of leadership and advocacy. They make me feel that I can contribute uniquely to something that can make the world better.

What projects are you working on now?

I am currently working on two projects. The first one is a sensor wearable, I call InduGlove, that can be used to train students in engineering schools, new employees in industries and automatically document their progress. It uses deep learning algorithms to recognize patterns and sends feedback to a supervisor.

My second project involves collecting images from Google Earth and Wind Data to predict and give insights about the exact location of water hyacinth, an aquatic plant, in Lake Victoria. Predicting the exact location of these weeds can be tricky as wind and other factors constantly move them to new locations. When we can predict and map exactly where they are despite the environmental factors, I think that would be a huge step towards eradicating them.

Tell us about a technology challenge you’ve had to overcome in a project?

I was building a digital clock in 2015, and I wanted to input a manual switch signal but every single press appeared like multiple presses. I kept on wondering what the problem could be for a long time. I later learnt that I needed to de-bounce the signal. With a de-bouncing circuit, I could tune it exactly to my requirements. The circuit could be adjusted to choose which pulse is rejected and which one is accepted.

I also found that I could record the signals and analyze them in real time using simple software routines. I could add a time delay after the first de-bounce and force the microcontroller to wait the same amount of time for the de-bouncing to stop and then continue with the program. The project is open source and can be found in my publication.

What trends do you see happening in technology in the near future?

The way I see it, I think Machine Learning & Artificial Intelligence, 5G & IoT, especially in terms of smart homes, and Self driving cars are definite trends for our future. I believes this is the year that AI-enabled decision systems will move from exploration and experimentation to enterprise adoption across a variety of industries including consumer applications.

I also see advances in 5G technology. 5G has been making the headlines. I think there will be more focus on it over time too because it will be the enabler of IoT and resulting interconnectedness of smart home technology. I think this is the year that 5G prototypes will start to take shape. I also think self-driving cars will be a big thing. These connected cars will be able to talk to each other and navigate on their own. With this ability, they could save millions of lives.

Developers and entrepreneurs should think about how they can leverage these technologies today.

How are you planning to leverage Artificial Intelligence or Deep Learning technologies in your work?

In my first project, I plan to use training and prediction algorithms in the Intel® Data Analytics Acceleration Library to predict the exact location of water hyacinth in Lake Victoria. I will use the factors that affect the movements of these weeds as my training data set. The prediction algorithm will then use the trained model to predict the exact location based on new data.

I also plan to use the k-Nearest Neighbors (kNN) Classifier in my InduGlove project to determine if a student is holding a tool correctly.

What are you looking forward to doing with Intel?

I want to work on three things with Intel: development, advocacy, and community. I plan to work on more Machine Learning and IoT projects, document them and share with others on how to build them. I also plan to demo my projects at trade shows and conferences.

When not building products, I love to teach. Sharing ideas is a very powerful aspect of my life. I believe education should be free and accessible to all. I plan to frequently attend events to reach as many developers as possible, connect with them and speak publicly about Intel’s cutting-edge machine learning, deep learning, and AI technologies. This includes both the large audiences and even smaller groups. Although I love presenting to large audiences, in some cases I find smaller groups much more intimate and it allows me to deliver highly relevant information tailored to that group.

I plan to host meaningful and impactful events in my community and highly encourage the developers in my community to join the Intel® Developer Mesh online community. I strongly believe that we can find solutions to our daily challenges and make remarkable contributions to our society through collaboration. This will create momentum and drive the success of adoption and integration of Intel architecture. It will also keep me connected to the communities that love technology as much as I do, especially ones that are focused around the Intel platform. I believe that the more people can learn how to make things, the more problems can be solved.

I enjoy writing technical articles on my blogs and producing digital video on my vlog. As an Ambassador, I will actively continue supporting developers by using my blogging and video editing skills, to constantly share my passion and experience by publishing high quality technical videos and tutorials, writing code samples, mentoring developers and much more. I’m also available for guest blogging and interviews upon request. I will focus on making the get started guide more clear and precise and provide remedies to common problems.

I frequently post on my social media accounts about AI and IoT. I will continue to do so and use the platform to answer developer questions. I will help Intel get a better understanding of my community’s culture, listening to developers’ feedback and being their voice to the Intel engineering teams.

As part of the Student Ambassador program, I look forward to connecting with thinkers, mentors, problem solvers and passionate people who are committed to bringing positive impact into the world. I hope this will help drive innovation and positively impact the lives of many people and communities.

How can Intel help students like you succeed?

By providing information and resources. I am extremely excited that Intel is providing direct access to its remote servers, access to early information, access to optimized technology, industry experts, and providing free software, tools, and libraries. This is a great step in helping us succeed. We also have a chance to join Developer Mesh, a developer social network that enables us to collaborate on projects, share research, and showcase our work.

Today we can easily organize our workshops as Intel is providing sponsorship funds to support the cost of hosting our events. This give us an opportunity to bring together likeminded student developers and build a greater developer community in Africa where we can experiment and learn together. It also enables us to have a voice and influence in driving AI forward.

What impact on the world do you see AI having? And do you see yourself as part of it?

In medicine, I can see AI being used to help doctors recognize disease and spare lives. I also observe AI empowering robots to help surgeons by passing them the necessary instruments during a medical procedure. AI software will track changes in health records to diagnose and analyze patients or caution doctors of potential risk factors and problems with medications designed.

I also see AI solving climate change. With more access to data, AI could distinguish patterns and utilize that data to think of answers for the world’s most concerning issues. Yes. I strongly believe that I will be part of this revolution. I will be involving myself more with companies, such as Intel, to do the three things that I love to do: development, advocacy, and community.

My current goal is to focus and enhance expertise in the area of IoT, Machine Learning & AI, Electronic Design & Prototyping and Cloud with special emphasis on deep learning. I also look forward to making 5G a reality, and enabling the IoT and AI to run over it. Although these areas present some exciting opportunities, I know it won’t be easy, but I am ready to adapt to any challenge and look forward for to it. With this, I believe I will be in a better position improve other people’s life and design projects that promote sustainable living.

Outside of technology, what type of hobbies do you enjoy?

I travel a lot and take random photos of places. It’s a good way to release daily stresses and frustrations accumulated during the week. I actively create, design and update changes to Google* Maps. I also review, approve or decline changes made by other map makers to keep Google Maps correct and up-to-date to the highest standards. By contributing to Google Maps, I positively impact the lives of many people and communities. I am focused on adding places like hospitals and bus stations. I am also passionate about verifying and correcting opening and closing hours of businesses. My contributions help everyone make better decisions about places to go. I am currently a level 4 Google* Local Guide.

I also make videos, especially for my events. I mix them with my love of art. There is a way in which all things are beautiful. I try that to make sure my videos exemplify a wabi-sabi aesthetic. My latest series of promotional videos are under Project Be There, where I encourage students and developers to attend my events. In the promos, I testify my dreams, obsession, ambitions, passion, ideologies, values and personal experiences.

I also teach. I have been involved in the Intel® She Will Connect Program for the past two years. In this program, I train a diverse groups of women on Computers and the Internet. The Intel® She Will Connect program uses digital literacy training, peer networks, and gender-relevant content to fix the Internet gender gap.

Want to learn more about the Intel® Student Developer Program for Artificial Intelligence?

You can read about our program, get a program overview, and we also encourage you to check out Intel® Developer Mesh to learn more about the various projects that our student community is working on.

Interested in more information? Contact Niven Singh

Originally published at software.intel.com on May 30, 2017.