Minor: Economics. Learning across borders, cultures, and economies.
Understanding power, policy, and how societies work.
Exploring how tech connects people and systems.
Minor: African Studies. Deep research into how people and technology shape each other.
Practical credentials that complement the research: agile delivery, project leadership, and the technical fundamentals underneath it all.
Helping teams work smarter with agile methods.
Planning, leading, and delivering real-world results.
Getting hands-on with the building blocks of tech.
Understanding how the internet really connects us.
Featured Project
An applied thread of my dissertation: community-driven language curation and translation work for Swahili, built with linguists and volunteers across East Africa.
Interdisciplinary research on AI, language, and digital labor, grounded in Africa and the Global South.
Hi! I'm Yinka, a Ph.D. graduate of the Information School at the University of Wisconsin–Madison, where I earned my Ph.D. in Information with a minor in African Studies.
I'm an interdisciplinary researcher, educator, and technologist interested in how technology shapes people, communities, and societies. My work sits at the intersection of AI, language and technology, digital labor, and technology and society, with a particular focus on Africa and the Global South. My doctoral research examined translators and digital labor in Kenya, exploring how digital platforms, computer-assisted translation tools, and AI are reshaping professional translation and the experiences of language workers.
Beyond academia, I am a UX research practitioner with experience consulting for sports and health startups and working as a Design Researcher at Symplicity, an education technology company in Virginia. I also build and collaborate on projects that explore how technology can better serve communities. Through Swahili Verse, I work to improve the representation of African language data in digital technologies and AI, with a particular interest in building more inclusive technologies and datasets.
I have taught and supported courses exploring digital inequality and access, data and algorithmic ethics and policy, human factors in information security, and the relationship between code, technology, and power. My teaching also examines how technology and society intersect within developing and emerging contexts, including questions of digital development, culture, inequality, and the social impact of emerging technologies.
Across my research, teaching, and applied work, I am interested in understanding technology not simply as a tool, but as something deeply connected to people, institutions, culture, and power.
I use tools like Copilot, Claude, and ChatGPT selectively, as part of a broader research workflow, not as a replacement for judgment. In academic and UX research work alike, AI helps me test alternative perspectives, explore early-stage ideas, and interpret complex information faster. It broadens the possibilities I consider; it doesn't decide my conclusions.
I remain fully responsible for the accuracy, interpretation, and ethical integrity of anything AI touches in my work. Every AI-assisted step gets verified against my own judgment, with awareness of the biases baked into training data and the systems behind them. AI supports my research and writing; it never stands in for the rigor the work requires.
Book chapters, peer-reviewed articles, and conference work on language technology, African studies, and information systems.
Courses taught at the Information School, University of Wisconsin–Madison.
Grounded in a constructivist approach, my teaching is organized around three interconnected principles: fostering collaborative learning, cultivating critical thinking, and building a culture of reflection, woven together with digital tools and interdisciplinary perspectives.
Download Teaching Philosophy (PDF)I turn real human behavior into product decisions through research that is inclusive, culturally grounded, and evidence-led.
Clarify user and business needs through stakeholder input and desk research.
Turn open questions into ones that guide real product decisions.
Mix qualitative methods like interviews and testing with quantitative ones like surveys and analytics.
Structured sessions with participants who reflect the real user base.
Patterns become personas, journey maps, and recommendations.
Share findings, track impact, validate with follow-up testing.
I study how people's behaviors, expectations, and contexts evolve, so product decisions stay grounded in what people actually need.
When people struggle with a product, I investigate the assumptions, systems, and design decisions around them before treating their behavior as the problem.
Five engagements across health tech, sports tech, career tech, language, and sales.
Care-ops platforms are dated and almost AI-free. That gap only pays off if the AI stays auditable.
Porchlight is a staff-first care-operations platform built for residential care facilities, starting with Arizona behavioral health residential facilities (BHRFs) and I/DD group homes. Most facilities juggle shift notes, medication records, incident reports, staff handoffs, and state-mandated compliance forms across disconnected tools: spreadsheets, ad hoc digital forms, verbal handoffs with no lasting record. Every undocumented gap is a compliance risk waiting to surface at audit time. This engagement was commissioned to give an honest read on the existing market before any feature got designed or built.
Question: Where do existing care-ops platforms fall short, and where's the real opening for Porchlight to differentiate?
The main stakeholders were Porchlight's founding and product leadership, who needed market evidence to validate or challenge early assumptions before committing engineering time. Product managers used the findings to sequence the roadmap. Design was a close second audience: the recommendation to prioritize a modern interface only holds up if it's grounded in a concrete account of what "dated" looks like across ten real competitors. Engineering leadership weighed in on feasibility, especially keeping AI-assisted features human-reviewable and auditable. Direct-care staff, house managers, and compliance officers weren't interviewed in this phase, but they're the ultimate beneficiaries of whatever gets built.
Competitive and market analysis was the right method. The question wasn't yet what users need specifically, but what's already been tried and where it fell short. I reviewed ten existing care-operations and resident-management platforms serving residential care and adjacent compliance-heavy sectors, scoring each against interface design, workflow coverage, and AI use. Findings fed into workflow mapping that turned competitive observations into buildable requirements.
Athletes weren't blocked by the interface. They were blocked by billing confusion.
Plyr is an athlete-profile platform built on a simple idea: athletes trying to get scouted don't have one credible, easy-to-build record of their performance. Plyr's core loop pulls stats, film, and achievements into one profile: build it, upload highlights, share one link, track who's viewing. Because Plyr is an active product, this research ran longitudinally instead of as a single round of interviews.
Question: How do athletes' expectations of the product change over time, and what should that change about the roadmap?
Plyr's founders were the most direct stakeholders and the primary audience for this research, since at an early-stage startup, roadmap and pricing decisions sit close to founder judgment. Product and design were the second core audience: deprioritizing interface polish in favor of billing clarity is a real reallocation of time, and needed to hold up on evidence. The fifteen club athletes who took part were a stakeholder group in their own right, representing the demand side of Plyr's target user base. Scouts, agents, and coaches sit further downstream as the platform's other core audience.
I directed a diary study paired with in-depth interviews across fifteen club athletes. Diary studies were the deliberate choice, since the question was about how expectations evolve and a single interview would only capture one moment. Interviews went deeper on behaviors the diary entries surfaced but couldn't fully explain. A competitive analysis benchmarked Plyr against other leading sports platforms.
Progress and recognition drove return visits nearly as much as the app's core features.
Symplicity's mobile app serves college users navigating career services: job and internship search, employer connections, career development resources. The mobile team noticed significant drop-off right after onboarding, followed by inconsistent return usage. The working theory was that low motivation, not missing functionality, was driving the drop-off, and this engagement tested that theory before committing design and engineering time to gamification.
Question: How might we design gamification elements that meaningfully increase engagement and retention without compromising usability or career development intent?
Product managers were the primary stakeholders, since the business case rested on retention and monetization tied to onboarding drop-off. UX/UI designers were the second core audience, translating behavioral findings into interface patterns. Front-end engineers were consulted on what was technically feasible to ship. The 150 participants, freshmen and sophomores navigating early career decisions, were the population this engagement aimed to represent accurately, though NDA terms limit how much specific input can be shared.
Primary methods were a survey and interviews, with A/B testing as a secondary method to pressure-test gamification concepts. I fielded the survey to 150 freshmen and sophomores, gathering data on usage habits and what drove or blocked engagement. Interviews added depth; responses went through thematic analysis, and A/B testing evaluated a shortlist of concepts directly. The screens developed alongside this research applied the Octalysis gamification framework across job discovery, XP and reward mechanics, streak tracking, and in-app messaging.
One "correct" translation broke trust. Native speakers expected dialect.
Existing digital Swahili glossaries and dictionaries are built almost entirely around lookup speed, a priority that flattens meaning, obscures regional variation, and gives users little basis to trust any single definition. This engagement was commissioned as foundational research to determine whether a genuinely different kind of Swahili language repository was worth building. It became the direct seed of what would later be built as Swahiliverse.
Question: How can a digital language platform provide clarity, trust, and contextual understanding for learners, native speakers, and translators in ways existing repositories do not?
Native speakers, language learners, and professional translators were interviewed not just as research subjects but as the people whose future participation the whole platform concept depended on. The broader stakeholder group included the initiative's founding team, who needed the research to validate whether a community-driven repository was worth building, and prospective content contributors and technologists who'd eventually build the platform. Swahiliverse didn't exist yet; the research itself became the platform's founding rationale.
I conducted twelve interviews across three groups: five native speakers, four learners, three professional translators, intentionally, since a repository designed around only one perspective risks missing what others need to trust and adopt it. In parallel, I ran a competitive UX analysis of six widely used Swahili glossaries, assessing layout, contextual examples, contributor transparency, and whether each tool supported exploration or only direct lookup.
Capture speed wasn't the real problem. Losing context was.
Fynn is a productivity and sales-enablement platform built to help sales teams capture, organize, and manage leads during trade shows and networking events, environments defined by short windows of attention and almost no time to properly document anything in the moment. This engagement was scoped to observe the real-world lead-capture workflow as reps actually experience it, and translate that into concrete guidance for Fynn's MVP.
Question: How can sales reps efficiently capture and manage leads without losing context in a fast-paced trade show environment?
My primary stakeholder was the product team, who commissioned this research to inform MVP scope and feature prioritization ahead of a build phase. Given the freelance, remote nature of the engagement, findings were delivered as structured research deliverables meant for internal decision-making. The sales representatives interviewed and observed on the trade show floor were the second, equally critical stakeholder group. Their managers and sales leadership sit as a further stakeholder layer, clearly implicated by the findings even if not directly consulted.
I combined interviews with three sales representatives with direct, live observation of their workflows on the show floor: watching lead capture and the exact moments where context visibly got lost, rather than relying solely on after-the-fact recall. A competitive analysis of existing lead-capture tools fed into an end-to-end journey map of the rep's experience, from initial contact through follow-up.
Thank you for looking through my work. ajibolayinka45@yahoo.com · More work at yinkaajibola.com
Fellowships, field research, conference acceptances, and dissertation milestones, in their own words, as they happened.
Reach out about research collaborations, speaking, or the Swahili Verse Project.
Pick whichever channel suits you best — I read and respond to all of them.
Translators as Gig Workers: Re-Iterating White Collar Work in the Era of Digital De-Professionalization in Kenya.
My engagement with data curation and translation challenges drew my attention to the linguistic and technological gaps shaping everyday translation practices. Over time, I observed that many translators operate within emerging gig-based, platform-mediated labor systems, highlighting a critical intersection between language work and digital employment. These insights shaped my research focus on the relationships among technology, language, labor, and translation, and led me to focus my research on Kenya. These experiences ultimately shaped the trajectory of my dissertation, "Translators as Gig Workers: Re-Iterating White Collar Work in the Era of Digital De-Professionalization in Kenya," which broadly explores how digital platforms are transforming the nature of gig work, particularly translation work in Kenya.
My research examines how platform-mediated systems are restructuring translators' employment models and redefining professional identity within the country's growing gig economy. I position translators as central actors in this ecosystem, analyzing how they navigate the uncertainties of digital labor while also probing issues in translation technologies, such as Computer-Assisted Translation tools, and the role of generative artificial intelligence in translation, particularly for low-resourced languages. Finally, I emphasize the importance of language, particularly African languages that remain underrepresented in global AI systems. Building on this commitment, I extend my work through an applied project focused on curating language data and developing resources for multiple Swahili dialects. This effort addresses a critical gap in computational linguistics by foregrounding cultural and linguistic diversity in dataset design.
Drawing on fieldwork in Kenya, the core chapters of my dissertation explore how gig translators navigate structural components and digital tools in their everyday work. Dorothea Kleine's Choice Framework, informed by Amartya Sen's Capabilities Approach, provides the theoretical grounding, highlighting how workers' well-being depends on their ability to exercise genuine agency within unequal digital labor environments.
Freelance translators in Kenya operate in a gig-based digital ecosystem. They juggle multiple platforms, clients, and deadlines while managing uncertain income and limited professional recognition. In the first study, I interviewed translators to understand their resource portfolios: a combination of skills, education, networks, infrastructure, and social support that enable them to survive and thrive in this environment.
Next, I explored the tools translators use, particularly Computer-Assisted Translation (CAT) systems. While these tools offer efficiencies such as translation memories and term databases, they often fail to capture local languages, idioms, and cultural nuances, limiting their usefulness in African contexts.
Finally, I examined Generative AI (GenAI) and its role in translation. Freelance translators in Kenya increasingly experiment with tools like ChatGPT. My experimental study revealed a mixed landscape: although AI often produces correct translations, it frequently struggles with idioms, contextual interpretation, and nuanced pragmatic meaning. For instance:
Observation: The entire sentence was translated directly, without taking context into consideration.
Mapping the intersections of freelance and gig translators, translation technologies, and AI has direct, practical implications for the applied world. By analyzing translators' resource portfolios, organizations can identify gaps in education, access to technology, and professional networks, enabling targeted interventions such as training programs, mentorship, or infrastructure support to improve income stability and career growth.
Insights from CAT tool analysis reveal limitations in supporting local languages and capturing cultural nuance, providing a clear roadmap for developers to design inclusive, context-aware translation technologies that better serve low-resource language communities. Similarly, examining human-AI collaboration highlights where skilled post-editing is required, informing the creation of efficient workflows that combine AI productivity with human expertise.
Collectively, these mappings enable policymakers, NGOs, and technology developers to make data-driven decisions that strengthen digital labor ecosystems, reduce socioeconomic disparities, and ensure that AI tools reflect linguistic and cultural realities, creating a more equitable and practical landscape for translators and technology users alike.