MOTIVATION LETTER The Mozilla Fellowship funds people who build technology that serves the public interest, not surveillance or shareholder extraction. I am an independent computational researcher from Nigeria, trained as a pharmacist, now building open-source tools for addiction neuroscience and drug safety. My research is pre-registered, my code is public on GitHub, and my models are designed for populations that pharmaceutical companies ignore. My primary research line is the Conjunctive Consolidation Threshold model, a tripartite pharmacological framework for preventing reward-memory encoding in addiction. The model couples dopaminergic reward-prediction error, NMDAR-dependent long-term potentiation, and affective contrast into a system of ordinary differential equations solved with RK45. I calibrated 14 free parameters using Bayesian MCMC with literature-elicited priors from an 1,847-record screen. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. The work is sole-authored and deposited on OSF and Zenodo. A co-authored paper is under review at Alcohol. I also built TOPOLOGIX, a protein-language-model classifier that predicts drug-resistance mutations from sequence alone. It achieves AUROC 0.804 on the Platinum benchmark, covering 100 percent of mutations compared to roughly 18 percent for structure-limited tools. This matters for African health systems where sequencing is cheaper and faster than crystallography. My neurocascade simulation engine couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics, passing 62 of 62 tests. My ergofluids project tested Koopman-operator methods for drug transport through tumor tissue and reported a failed pre-registered gate directly rather than reframing it. These projects share a method: transparent, reproducible, falsifiable computational science. I do not hide negative results. I do not claim impact I cannot measure. That is the public-interest approach Mozilla exists to support. I am enrolled in the M.Sc. Digital Health at Hasso Plattner Institute in Potsdam, starting winter 2026. I hold a B.Pharm from the University of Ibadan and am licensed by the Pharmacists Council of Nigeria. I currently work as National Product Manager at Synthcare. My endorsers include Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. The Mozilla Fellowship would allow me to dedicate twelve months to extending the CCT model into a deployable clinical decision-support tool for addiction treatment in low-resource settings, and to open-sourcing the full neurocascade simulation pipeline. I will build these in the open, document them for non-specialist clinicians, and test them against real behavioral data from collaborators in Nigeria. I am applying to build something that works for people who have been systematically excluded from computational psychiatry. RESEARCH STATEMENT My research program addresses a single question: can we build computational models of brain function that are transparent enough to trust, cheap enough to deploy, and specific enough to guide clinical decisions in addiction and drug safety? The CCT model is my central project. It formalizes a hypothesis about how reward memories consolidate during addiction and how pharmacological intervention at three coupled axes can prevent that consolidation. The model is a system of ODEs with 14 parameters, each constrained by published experimental data. I calibrated it with Bayesian MCMC using PyMC and the DEMetropolisZ sampler. The posterior distributions confirmed all five pre-registered hypotheses. The next step is to fit the model to real human behavioral data from cue-reactivity and reconsolidation paradigms. I have identified collaborators at the University of Ibadan who run these experiments. The Mozilla Fellowship would fund the data-sharing infrastructure and the computational time needed to complete this fit. My secondary line is protein-level drug safety. The TOPOLOGIX project uses ESM-2 protein-language-model embeddings combined with Morgan fingerprints and a Random Forest classifier to predict drug-resistance mutations. It achieves AUROC 0.804 on the Platinum benchmark and 0.634 on SKEMPI 2.0. It covers all mutations, unlike structure-based tools that cover only 18 percent. This is directly relevant to antimicrobial resistance surveillance in Nigeria, where the Genomic Surveillance of Antimicrobial Resistance unit at the Nigerian Centre for Disease Control sequences pathogens but lacks tools to predict resistance from sequence alone. I have built a DuckDB-based pipeline that ingests resistance data and runs TOPOLOGIX predictions. The pipeline is open-source and deployable on a laptop. My third line is simulation infrastructure. The neurocascade engine couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics to behavioral readout. It passes 62 of 62 unit tests. The circuit-layer parameters are currently illustrative because I have not yet fit them to real behavioral data. That is the next milestone. The ergofluids project tested whether Koopman-operator methods with Mori-Zwanzig memory kernels can model drug transport through tumor tissue. The first real-data gate failed its pre-registered criterion. I reported that result directly rather than reframing it as a success. That is the standard I hold myself to. All code is on GitHub. All preprints are on OSF and Zenodo. All analyses are pre-registered where possible. I do not publish results I cannot reproduce. SHORT ESSAY: PUBLIC INTEREST AND OPEN TECHNOLOGY Mozilla defines the public interest as technology that empowers people, respects privacy, and remains open to scrutiny. My work meets that definition by construction. The CCT model is built on open-source Python libraries and deposited on OSF. Any researcher with a laptop and an internet connection can reproduce my results. The model is designed for addiction treatment in settings where expensive neuroimaging is unavailable. A clinician in a Nigerian teaching hospital can use the model to simulate the effect of a pharmacological intervention on reward-memory consolidation using only a patient's medication history and a brief behavioral assessment. That is not a future claim. The ODE system runs in under two seconds on a standard laptop. TOPOLOGIX is open-source and requires no structural biology expertise. A public-health officer in an African genomic surveillance lab can feed a pathogen sequence into the classifier and get a resistance prediction in minutes. The alternative is sending samples abroad for crystallography or relying on tools that cover only a fraction of known mutations. That is a concrete equity gap, and my tool closes it. I also practice what Mozilla preaches about transparency. When the ergofluids project failed its first real-data gate, I published the negative result. When the hERG cardiotoxicity study found that topological features do not beat a plain descriptor baseline, I reported that. The addiction neuroscience literature is full of models that look good on synthetic data and fail on real data. I am building a reputation for reporting the failures. The Mozilla Fellowship would allow me to formalize these practices into documentation, tutorials, and deployment guides for non-specialist users in low-resource settings. That is the public-interest work I am already doing, at scale. SHORT ESSAY: LEADERSHIP AND TRACK RECORD Leadership in computational research does not require a lab or a faculty appointment. It requires producing work that others can build on, and doing so consistently and transparently. I have produced five sole-authored preprints and one co-authored paper under review at Alcohol. I have built four independent DuckDB-based ingest-to-analyze pipelines across life sciences, technology, and social science domains. I maintain a self-hosted local LLM serving stack with on-demand model swapping. I have deployed production systems with systemd, Caddy TLS, and automated backup and disaster recovery. I have worked as a clinical pharmacist, a bioinformatics researcher at the Genomic Surveillance of Antimicrobial Resistance unit, and a national product manager at Synthcare. My endorsers include Kent Berridge at the University of Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. These are researchers whose work I have built on and who have read my preprints. I did not ask for endorsements from people who do not know my work. I am 29 years old, Nigerian, and enrolled in an M.Sc. Digital Health program at Hasso Plattner Institute. I have no PhD and no faculty position. I have built a research program that produces testable, falsifiable, open computational models. That is leadership by output, not by title. The Mozilla Fellowship would give me the time and resources to extend this work into deployment. I will train two to three early-career researchers in Nigeria on the CCT model and TOPOLOGIX pipeline during the fellowship year. I will publish all training materials openly. I will report progress and failures publicly. CHECKLIST - [ ] Motivation letter, 500 words maximum, tailored to Mozilla Fellowship - [ ] Research statement, 600 words maximum, describing CCT model, TOPOLOGIX, neurocascade, and ergofluids - [ ] Short essay on public interest and open technology, 350 words maximum - [ ] Short essay on leadership and track record, 350 words maximum - [ ] CV or resume, two pages maximum, listing education, employment, publications, preprints, and skills - [ ] Two letters of recommendation, one from a research collaborator or endorser - [ ] Proof of enrollment in M.Sc. Digital Health at Hasso Plattner Institute - [ ] Proof of B.Pharm degree and PCN pharmacist license - [ ] Links to ORCID profile, GitHub repository, and personal website - [ ] Links to all preprints on OSF and Zenodo - [ ] Links to code repositories for CCT model, TOPOLOGIX, neurocascade, and ergofluids EDITOR NOTES - Eligibility risk: the Mozilla Fellowship typically requires applicants to be based in a country where Mozilla operates or to have a clear connection to the open internet movement. Eniola is Nigerian and enrolled in a German university. Confirm that Nigeria is an eligible country for this specific fellowship cycle. The URL provided is a third-party aggregator, not the official Mozilla page. Verify the official deadline and eligibility criteria at foundation.mozilla.org. - Fact to verify: the co-authored paper in Alcohol is listed as under review. If it is accepted or rejected by the deadline, update the status. If it remains under review, state that clearly. - Gap to fill: the application materials do not specify a project title or a concrete 12-month work plan. The Mozilla Fellowship typically asks for a project proposal with milestones. Draft a one-page project plan titled "Open Computational Models for Addiction Treatment and Drug Safety in Low-Resource Settings" with quarterly milestones, deliverables, and a budget estimate for travel and computing costs. - Gap to fill: the application does not mention any prior relationship with Mozilla or Mozilla-affiliated projects. If Eniola has used Firefox, contributed to Mozilla open-source projects, or attended Mozilla events, add that detail. If not, consider whether to mention Mozilla's mission explicitly in the essays as already done. - Personal detail needed: the application should include Eniola's current contact information, including email address and phone number, and a professional headshot if the submission portal requires one.