Sakayo Toadoum Sari

<!-- Animated header -->

<!-- Typing animation -->

<!-- Social badges -->

---

$ whoami

python
class Sakayo:
    role        = "PhD Candidate in AI @ CRIL-CNRS (UMR 8188), Université d'Artois"
    thesis      = "Intertextuality & Causality: Towards a Semantic Analysis of Discourses"
    also        = "Lead NLP @ PlotWeaver — voice AI for African languages (NVIDIA Inception)"
    focus       = ["Discourse Semantics", "Causal Reasoning", "Formal Explainability (XAI)"]
    languages   = {"native": ["Mbay", "French"], "fluent": "English",
                   "shipped": ["Ngambay", "Hausa", "Yoruba", "Wolof", "..."]}
    teaching    = "Algorithmics @ Université d'Artois · 80+ students trained across Chad & Cameroon"
    mission     = "Building language tech for communities the industry forgot"

---

🔭 Current Work

| Project | What | Status |
|:--------|:-----|:------:|
| 🎓 PhD Thesis | Intertextuality & causal reasoning in discourse — bridging NLP with SAT/formal methods |  |
| 📖 IntertextMigra | Detecting experiential intertextuality in migration narratives |  |
| 🇫🇷 EvalLLM RAG | Solo end-to-end French RAG pipeline (BM25 + dense + reranking) |  |
| 🔍 Discourse encoders, formal XAI & more | Lightweight discourse-relation models, minimality-guaranteed probing, interpretable rule learning |  |
| 🎙️ PlotWeaver | Production TTS/ASR for Hausa & Yoruba on AWS SageMaker |  |

---

📚 Selected Publications

🆕
IntertextMigra: Experiential Intertextuality in Migration Narratives — SIGDIAL 2026 (accepted)

🆕
French RAG under Strict Constraints — CORIA-TALN 2026 (accepted)

📄
Pretraining Strategies for Low-Resource MT — AfricaNLP 2025 ↗

📄
Ngambay–French Neural Machine Translation — NLP4TIA @ RANLP 2023 ↗

📄
MasakhaNEWS: News Topic Classification for African Languages — IJCNLP-AACL 2023 ↗ (most-cited work)

📄
AfriMTE & AfriCOMET — MT Evaluation for Under-resourced African Languages, NAACL 2024 ↗

📄
Error Correction DNNs for Wildlife-Vehicle Collision Data — MDPI Mathematics ↗

> 📎 Full list & citation count → Google Scholar · toadoum.github.io

---

🗺️ Journey So Far

2026 ──── PhD Year 2 @ CRIL-CNRS — first papers accepted (SIGDIAL, CORIA-TALN)
  │       Teaching Algorithmics @ Université d'Artois
  │       Founder & CEO @ PlotWeaver — Hausa/Yoruba voice AI in production
  │
2025 ──── PhD in AI begins @ CRIL-CNRS, Université d'Artois (ANR HYCI project)
  │       UNICEF Consultant (RCCE) — Chad
  │       ML Engineer @ CDIAL — African ASR/TTS
  │
2023 ──── NLP Engineer @ Batazia — multilingual product NLP
  │       Summer Research Intern @ EPFL — polarization detection
  │       Lead Trainer @ Tec4Tchad — Python, ML, Git (40+ students)
  │       MSc Machine Intelligence (AMMI) — Google & Meta Scholar
  │
2022 ──── Project Lead ML Intern @ STEM-Away
  │
2020 ──── MSc Industrial Mathematics — AIMS Cameroon (MasterCard Foundation)
  │
2019 ──── MSc Physics — Université de N'Gaoundéré
  │       Web Dev & ML Intern @ Limbe City Council
  │
2017 ──── BSc Applied Physics — the beginning

---

🛠️ Tech Stack

Languages & Frameworks

!Python
!PyTorch
!HuggingFace
!FastAPI
!Django

Infrastructure & Tools

!AWS
!SageMaker
!Docker
!SLURM
!Git
!LaTeX

Specializations

!NLP
!XAI
!ASR
!TTS
!Machine Translation
!Low--Resource Languages
!RAG

---

🏆 Highlights

🎓 Google & Meta Scholar (AMMI) · 🎓 MasterCard Foundation Scholar (AIMS) · 🚀 NVIDIA Inception (PlotWeaver)

📝 7+ Publications · 🌍 5+ African Languages in production NLP systems · 🎙️ Voice AI serving real users

🔬 Reviewer: Masakhane African Languages Hub, IGI Global · 👨‍🏫 Trained 80+ students across Chad & Cameroon

📺 Sakayo Tech — AI education for francophone African students · 🤝 Mentorship program for 20 young Chadian innovators

---

📊 GitHub Activity

&nbsp;&nbsp;

---

📫 Let's Talk

Building something for underserved languages? I'd love to hear about it.


Collapse