🔹 Jay Karippacheril Jacob — MSc in Computer Simulation in Science (GPA 1.9), specializing in Computational Fluid Dynamics.
🔹 Currently working on my Master Thesis: Acceleration of Matrix Sign Functions — focusing on algorithm design, numerical stability, and performance optimization.
🔹 Passionate about Machine Learning, Deep Learning, and Data Engineering: building reproducible pipelines, training models, and deploying lightweight ML prototypes.
🔹 Experienced in simulation code optimization (C/C++/Fortran/MPI), numerical methods, and end-to-end ML experiments with TensorFlow/Keras.
🔹 Open to collaborating on Computational Fluid Dynamics, Deep Learning, and Software Engineering projects.
🔹 Actively improving skills in Python libraries, SQL, PySpark, and cloud/parallel computing techniques.
🔹 Fluent in English (C2) and intermediate German (B1 → B2 preparation).

Software & Simulation:

• Simulation code optimization in C/C++/Fortran with MPI/OpenMP.
• Computational Fluid Dynamics and numerical linear algebra algorithms.

Machine Learning & Data Science:

• End-to-end TensorFlow/Keras experiments: CNN image classification, LSTM time-series forecasting.
• Data pipelines (ETL) with Python, SQL, pandas, and reproducible notebooks.
• Dashboarding & reporting using Matplotlib, Plotly, Streamlit/Dash.

Selected Projects:

• 42 Heilbronn: low-level C projects emphasizing memory safety, parsing, concurrency, and algorithmic optimization.
• ML Prototypes: model training, feature engineering, cross-validation, inference scripts, lightweight serving.
• Forschungszentrum Jülich Internship: numerical feature implementation, profiling, mixed-precision optimization, and reproducible pipelines.

Domain Knowledge: predictive maintenance, anomaly detection, quality monitoring, fault detection, ETL processes, backend APIs.

• Contribute to Deep Learning / AI applications in industrial and scientific settings.
• Apply skills in numerical simulation, ML, and software engineering to real-world problems.
• Expand expertise in HPC, distributed systems, and scalable ML pipelines.