Curriculum vitae

Gabin Maxime Nguegnang

AI Research Scientist & AI/ML Engineer

Munich, Germany · gmnguegnang@gmail.com

Experience

AI Research Consultant: LLM reasoning and agentic AI evaluation

Outlier AI and Mercor · Remote, Germany

Oct 2025 – present
  • Find where a language model's step-by-step reasoning breaks on hard STEM problems and write the correction, which becomes the training signal (RLHF) for the next, more accurate version.
  • Grade AI agents on tool choice and on staying on track across long tasks (ReAct-style reasoning, function calling, long-context management), and settle the ambiguous cases against a written rubric.
  • Design the scoring rubrics and lead peer review, so prompt engineering stays consistent across a distributed group and hard mathematical content turns into training data the team can trust.

Doctoral research and teaching assistant

Ludwig Maximilian University of Munich · Munich, Germany

Apr 2024 – Sep 2025
  • Established the conditions under which stochastic gradient descent is guaranteed to converge, showing engineers how aggressively they can train before a run falls apart.
  • Delivered an explainable-AI (XAI) project that flags failures early in industrial power plants and gives the reason behind each alert. Reached F1 0.99 with 6 ensemble learners and SHAP, and held fairness across sites (Disparate Impact Ratio 0.95).
  • Taught postgraduate courses in optimization, machine learning, deep learning, and data science, and defended a disputed proof step through to acceptance by the Springer Nature editorial board.

Doctoral research and teaching assistant

RWTH Aachen University · Aachen, Germany

Sep 2019 – Mar 2024
  • Derived practical learning-rate conditions for gradient descent on deep networks that do not shrink exponentially as the network gets deeper, removing a limit that kept the earlier theory out of practical use.
  • Backed the theory with large-scale PyTorch training runs on high-performance computing infrastructure, spreading the workload across nodes to cut the runtime of each experiment.
  • Taught continuous optimization, mathematics of data science, and higher mathematics, and built course materials that made advanced theory concrete for engineering students, within a research group of 9 nationalities.

Machine learning intern

Group One Holding Company · Limbe, Cameroon

Jun 2018 – May 2019
  • Analyzed telecom fuel-consumption data to pinpoint the root cause of fuel loss and compared 4 machine learning models (Gradient Boosting led at 98% Nash efficiency).
  • Deployed the winning model as a Flask web application with a monitoring dashboard, securing 84,617 liters of fuel.
  • Automated log ingestion, cutting reporting time from days to seconds, and tracked API success rates and processing latency on the dashboard.
  • Mapped operational constraints with base-station technicians, then presented the findings, the model results, and the case for rollout to company managers and the operations director.

Education

PhD, Applied Mathematics: deep neural networks optimization

LMU Munich and RWTH Aachen University, Germany

Convergence analysis for training deep neural networks. Published in Springer Nature (2024) and SampTA 2023.

Sep 2019 – Sep 2025

Co-operative MSc, Industrial Mathematics: machine learning

African Institute for Mathematical Sciences (AIMS), Limbe, Cameroon

Mastercard Foundation Scholars Program award, 2017.

Aug 2017 – Feb 2019

MSc, Applied Mathematics: dynamical systems and modeling

University of Yaoundé I, Cameroon

Optimization and numerical analysis.

Oct 2016 – Nov 2018

BSc Honours, Applied Mathematics

University of Douala, Cameroon

Oct 2014 – Jul 2016

Publications

  1. Convergence of gradient descent for learning linear neural networks Springer Nature: Advances in Continuous and Discrete Models, 2024
  2. Analysis of stochastic gradient descent for learning linear neural networks SampTA 2023
  3. Anomaly detection in power generation plants using machine learning and neural networks Applied Artificial Intelligence, 2020
  4. Predicting fuel consumption in power generation plants using ML and neural networks ICECET 2021

Skills

Programming and data

Python (pandas, NumPy, SciPy, scikit-learn) · PyTorch · SQL · PostgreSQL · pgvector · FAISS · FastAPI · Uvicorn · Pydantic · Flask · REST APIs · Bash

Agentic AI, LLMs, and RAG

AI agents · Multi-agent orchestration · LangGraph · LangChain · MCP · DSPy · RAG · CRAG · ReAct · Chain-of-Thought · Long-context management · Hugging Face (Transformers, TRL, PEFT) · LoRA · QLoRA · SFT · RLAIF · DPO · RLHF

Evaluation and tracking

Evaluation framework design · Live evals · Failure-mode analysis · Root-cause analysis · Red-teaming · Rubric-based evaluation · Preference labeling · Quality gates · Recall@k · ALCE · RAGAS · Prometheus 2 · MLflow

Machine learning and research

Deep learning · Statistics · Non-convex optimization · Convergence analysis · Bayesian optimization · Experimental design · Gradient Boosting · Random Forest · SVM · Anomaly detection · Explainable AI (SHAP) · NLP · Operations research solvers

MLOps, DevOps, and cloud

Docker · GitHub Actions · GHCR · Kubernetes · Azure AKS · AWS · Google Cloud Vertex AI · OpenTelemetry · LangSmith · Git · pytest · Distributed training · HPC · Agile (Scrum)

Certifications

  • Advanced agent coding Outlier AI · Jan 2026
  • Advanced prompt engineering Outlier AI · Dec 2025
  • Generative AI with large language models DeepLearning.AI / Coursera · Oct 2024
  • Machine learning engineering for production (MLOps) specialization DeepLearning.AI · May 2024

Memberships and awards

  • Munich Center for Machine Learning (MCML), member Jun 2024 – present
  • Community of Computational and Mathematical Methods in Data Science, member Dec 2019 – present
  • Mastercard Foundation Scholars Program award Aug 2017

Languages

  • French Mother tongue (C2)
  • English Proficient (C1)
  • German Professional working proficiency (B2+)