Skip to content
View empresst's full-sized avatar

Block or report empresst

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. 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
empresst/README.md

Hi, I'm Fatema Tamanna

I'm a freelance AI engineer and ML/DL researcher. My research is on deep learning for time series and sequential data: hybrid forecasting models, and how sequence models behave and can be updated when their data changes over time. I also build AI agents and LLM applications that run in production.

Research

  • Hybrid Deep Learning Approach for Load Profile Prediction, STI 2024. BiLSTM/BiGRU/TCN with self-attention and an XGBoost meta-learner for smart-grid load forecasting. Paper · Code
  • Enhancing Prediction Accuracy using Ensemble Deep Learning Methods on Time Series Data, ICISET 2024. Paper · Code
  • Audit Before You Update: recording artifacts and the limits of bounded model updates in MIMIC-IV ICU prediction. Preprint and code: DataAudit
    • Companion study on explanation stability when part of the model is frozen: bounded_update

Live demos

  • Kotha AI: Bangla voice receptionist that books real clinic appointments over a call
  • AI Messenger: chat with a family member's AI twin (demo logins in the repo; the free server can take a minute to wake up)

Tools I use

Python, PyTorch, TensorFlow, scikit-learn, XGBoost, OpenCV, Hugging Face, LangChain, LiveKit, FastAPI, Next.js, PostgreSQL

Pinned Loading

  1. Java_AndroidApp Java_AndroidApp Public

    Android app for maternal health in Bangladesh: trimester diet plans, symptom-based disease prediction and reminders

    Java 8

  2. ICISET ICISET Public

    Official code for our ICISET 2024 paper: BiLSTM + BiGRU + TCN ensemble with an XGBoost meta-learner for time series forecasting

    Jupyter Notebook

  3. STI STI Public

    Code for our STI 2024 paper: hybrid BiLSTM/BiGRU/TCN with self-attention and XGBoost for load profile prediction

    Jupyter Notebook

  4. empresst.github.io empresst.github.io Public

    My portfolio website

    HTML 5

  5. PneumoniaDetectionExplainableAI PneumoniaDetectionExplainableAI Public

    Chest X-ray pneumonia detection (PyTorch, DenseNet-121) with patient-level splits and external validation on 6,000 adult RSNA X-rays: AUROC 0.94 vs normal, 0.80 vs all other findings. Includes cali…

    Jupyter Notebook

  6. AI_messenger AI_messenger Public

    Family messenger with an AI twin: privacy-filtered RAG over journals and chats, with a retrieval benchmark (top-2 accuracy 53% → 98%, zero privacy leaks)

    Python