Skip to content
View mamady1999's full-sized avatar

Block or report mamady1999

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
mamady1999/README.md

👋 Hi, I'm Mamady Konaté

Data Scientist | Seeking PhD in Healthcare AI | Deep Learning & Time Series

🇬🇳 Guinea → 🇸🇳 AIMS Senegal 📧 mamady.konate@aims-senegal.org • 🔗 LinkedIn


🎯 What I Do

I'm focused on applying ML to healthcare challenges—especially for places where access to good healthcare is limited.

Right now:

  • 💼 Developing mobile apps for medical claims at Teliya Santé (health insurance, Guinea)
  • 🔬 Applying deep learning to time series forecasting with uncertainty quantification
  • 📊 Training models that know when they don't know (because in healthcare, that matters)

Recently:

  • 📝 Submitted thesis work to Frontiers in Environmental Science (under revision)
  • ✅ Earned 4 WorldQuant University certifications in deep learning (RNNs, LSTMs, GANs, VAEs)
  • 🎓 Master's (Distinction) from AIMS Senegal with focus on uncertainty quantification

🔥 Why Healthcare AI?

I'm not just interested in ML—I'm obsessed with applying it to medicine.

Growing up in Guinea, I've seen firsthand the challenges of limited healthcare access and infrastructure. This drives my focus on building AI systems that work in real-world conditions—not just in well-funded hospitals with perfect data.

My mission: Build AI systems that are trustworthy and practical—systems that work in clinics with limited resources, for populations that are often overlooked in mainstream AI research.


🛠️ Tech I Work With

ML/DL: PyTorch • TensorFlow • scikit-learn
Sequence Modeling: RNN • LSTM • GRU • Transformers
Generative AI: GANs • VAEs • Autoencoders
Time Series: ARIMA • N-BEATS • N-HiTS
Computer Vision: YOLO • Transfer Learning
Languages: Python • SQL • Kotlin
Tools: Docker • Git • MLflow • Jupyter • VS Code


🚀 What I've Built

🏥 Flood Prediction with Uncertainty Quantification

Master's Thesis | AIMS Senegal

Built deep learning models to predict water levels in the Senegal River (Bakel station).

  • 📊 Analyzed 21,000+ daily observations (1960-2018)
  • 🧠 Compared ARIMA vs. neural architectures (LSTM, N-BEATS, N-HiTS)
  • 📈 Developed residual-bootstrap prediction intervals for uncertainty
  • 📝 Publication under revision at Frontiers in Environmental Science

Why it matters: Climate resilience for communities vulnerable to flooding.

📱 Medical Claims Validation App

Teliya Santé | Current

Kotlin Multiplatform mobile app for medical advisors to validate insurance claims.

  • 🏥 Healthcare tech improving insurance operations in Guinea
  • 📱 Cross-platform (Android/iOS)

👁️ Computer Vision for Logistics

Logidoo | 2024-2025

YOLO-based prototype for automated parcel dimension analysis.

  • 📦 Built end-to-end ML pipeline (data → training → deployment)
  • 🐳 Deployed with Docker

📊 Customer Behavior Analysis

MTN Guinea | 2022

Data analysis and geolocation support for telecom operations.

  • 📈 Analyzed behavioral data for retention strategies
  • 🗺️ Provided geolocation support for law enforcement

🎓 Education & Training

M.S. in Mathematical Sciences (Distinction)
AIMS Senegal, 2023-2025

B.S. in Software Engineering
Gamal Abdel Nasser University, Guinea, 2016-2020

Additional Training:

  • Cambridge Centre for AI in Medicine Summer School (September 2025)

WorldQuant University Certifications:


💭 What Drives Me

I don't just want to build models that perform well on benchmarks. I want to build systems that actually help people.

I'm drawn to problems where:

  • The stakes are high (healthcare, climate)
  • The data is messy (real world)
  • The resources are limited (most of the world)

I love understanding how things work "sous le capot" (under the hood). Not just applying methods, but really grasping the fundamentals.


📫 Let's Connect

Open to:

  • PhD opportunities in Healthcare AI
  • Research collaborations on time series + uncertainty quantification
  • Projects with real-world impact

Reach me:


"Building AI systems that work for everyone, not just the well-resourced."

Popular repositories Loading

  1. U.S.-Medical-Insurance-Costs U.S.-Medical-Insurance-Costs Public

    The purpose of this project is to analyze a csv file called insurance.csv, in order to have clear information about the cost of insurance in the United States according to several criteria represen…

    Jupyter Notebook

  2. Constellations Constellations Public

    Jupyter Notebook

  3. Capstone-Project-Netflix-Data Capstone-Project-Netflix-Data Public

    Python

  4. Life-Expectancy-and-GDP Life-Expectancy-and-GDP Public

    Jupyter Notebook

  5. Biodiversity-in-National-Parks Biodiversity-in-National-Parks Public

    Jupyter Notebook

  6. OKCupid-Date-A-Scientist OKCupid-Date-A-Scientist Public

    In recent years, there has been a massive increase in the use of dating apps to find love. Many of these apps use sophisticated data science techniques to recommend possible matches to users and op…

    Jupyter Notebook