Coding2 min read
yield in Python
Understanding the yield keyword in Python — how generators work, when to use them, and their key advantages over regular functions.
AI RESEARCHER · ENGINEER · EDUCATOR
Salomon Kabongo
I build, research, and teach artificial intelligence.
My work spans foundation models, multimodal AI, computer vision, document intelligence, scholarly information extraction, and technology for African languages. Through Dr. Kabongo, I teach practical AI in English, French, Lingala, Tshiluba, and Swahili.
Research
From understanding AI to building with it
I hold a PhD in Computer Science from Leibniz Universität Hannover. My research examines how AI systems find, connect, and evaluate evidence across language, documents, images, and scientific literature.
I connect peer-reviewed research with real-world engineering across foundation models, multimodal AI, computer vision, document intelligence, and reliable evaluation. That work includes an issued U.S. patent in image privacy and de-identification.
Research and education are part of the same project: useful AI knowledge should be rigorous, practical, and available beyond English. Through Dr. Kabongo, I teach in English, French, Lingala, Tshiluba, and Swahili.
That conviction also guides my work on DarAkili, a STEM education platform where African professionals publish free mini-courses in their own languages.
Research discovers what is possible. Engineering tests it in the real world. Teaching makes the knowledge travel.
Research interests
Foundation models & LLMs
How models use context, retrieve evidence, reason across tasks, and fail in practice.
Multimodal AI & computer vision
Systems that understand and connect images, video, text, and other signals.
NLP & information extraction
Turning documents and scientific literature into structured, useful knowledge.
Evaluation & reliable AI
Testing model quality, limitations, and fitness for consequential real-world use.
African-language technology
Data, models, and community-led methods for multilingual and low-resource settings.
AI education
Making modern AI understandable and useful across languages and experience levels.
Selected record
- PhD
- Computer Science
- Leibniz Universität Hannover · completed November 2025
- U.S. Patent 12,613,996
- Image privacy & de-identification
- Issued April 28, 2026
- Best Paper
- First-author research recognition
- ICADL 2021
- Applied AI
- Research to real-world systems
- Experience across multimodal, document, and generative AI
- Former board member
- Masakhane Research Foundation
- Governance and partnerships for African-language research
- 5 languages
- AI education
- English · French · Lingala · Tshiluba · Swahili
Video
Learn AI with Dr. Kabongo
Research explained, practical tools demonstrated, and honest conversations about modern AI — across languages and levels of experience.
Masakhane MT. How to get started ?
@DrKabongo
Poster 21 Black in AI Workshop @ NeurIPS 2021 by Salomon Kabongo
@DrKabongo
Hello World de l’IA | Machine Learning From Scratch: Apprendre à un ordinateur à additionner
@DrKabongoFR
Hello World ku AI mu Tshiluba | Machine Learning From Scratch -- Kulongesha Ordinatɛr Addition
@DrKabongoTshiluba
Hello World to AI | Machine Learning From Scratch: Teach a Computer to Add
@DrKabongo
Amazing interview with a mathematician - Dimby from Madagascar.
@DrKabongo
Channels
Four channels, education in five languages
YouTube channels are available in English, French, Lingala, and Tshiluba. Educational content and access on this site also include Swahili.
English
@DrKabongo
Research walkthroughs, practical AI, and clear explanations of foundation models, multimodal systems, and technical careers.
10 videos
SubscribeFrançais
@DrKabongoFR
L'IA expliquée en français, pour la francophonie africaine et mondiale — recherche, carrière et pratique.
1 video
SubscribeLingála
@DrKabongoLingala
AI, research and technology explained in Lingala — for Kinshasa, Brazzaville and the diaspora.
First video coming soon
SubscribeTshiluba
@DrKabongoTshiluba
AI, research and technology explained in Tshiluba — because a technical idea should not require a colonial language first.
1 video
SubscribeSpeaking
Talks, lectures and panels
I speak in English or French to research, technical, and general audiences about modern AI, applied systems, African-language technology, and technical careers.
What modern AI can — and cannot — do
A clear account of current capabilities, common failure modes, and how to evaluate claims without hype.
Multimodal and applied AI
How language, vision, video, and documents come together in useful systems — from research choices to real-world trade-offs.
African-language technology
What it takes to build AI with and for multilingual African communities, including data, evaluation, access, and participation.
Research and technical career development
Practical lessons on finding collaborators, publishing strong work, growing technical judgment, and building a durable career.
Writing
Writing
Coding2 min read
Circular Queue: Python Implementation
A clean Python implementation of the circular queue (ring buffer) data structure that beats 94% of LeetCode submissions.
Mathematics2 min read
Mutually Exclusive vs. Collectively Exhaustive
A quick primer on two foundational concepts in probability theory that are often confused — mutual exclusivity and collective exhaustiveness.
Contact
Work with me
Two focused ways to work together, with clear expectations and an email-first process.
Speaking invitations
Talks, lectures, and panels in English or French for research, technical, and general audiences.
Selective AI advisory
Bounded consultations, written reviews, and workshops for clearly defined AI questions.
Join the community
Two places to follow the work, ask questions and meet others learning the same things.
WhatsApp channel
Short updates when a new video or article goes out. Broadcast only — no group noise.
Follow the channel
Discord server
Where the conversation happens: questions, study threads, and help from people further along.
Join the server