Welcome to my GitHub! I am Brian, a bald, bearded South African postgraduate scholar with a deep fascination for Big Data, high performance and scientific computing. I am passionate about solving the scientific world's most challenging problems while absorbing any and all knowledge along the way. My academic journey is a tapestry of mathematics, radio astronomy, statistics and computer science.

• 📧 Email: brianallisterwelman@gmail.com
• 🔗 LinkedIn: /in/brianwelman
• 🌐 ORCID: 0000-0002-0254-5173
• 📸 Instagram: @brianwelman

• 🔭 Previously worked on: Just finished my masters in radio interferometry, focusing on developing a fast, scalable and sophisticated prototype calibration framework called Kalman Filtering and Smoothing in Calibration (KalCal). My thesis describes the theoretical underpinnings of the framework and the excellent results from a simulated calibration experiment to justify further development.
• 🖥 Currently working on: A novel real-time or "on-line" calibration software suite, written in Python and Rust, called the On-line Radio Interferometric Calibration Suite (ORCa). It forms the basis of future doctoral research into unprecedented real-time calibration technologies for modern radio interferometers, which starts the path to a "world-first" radio interferometer image preview system.
• 🎓 Learning and Development: Mastering the Rust programming language and associated best software development practices. Additionally, I am advancing my Bayesian statistics knowledge to solve parallel and distributed computing limitations of Bayesian filters and smoothers.
• 👥 I’m looking to collaborate on: Any open-source projects! I love giving back; programming and problem-solving are fun in any form. My expertise lie in mathematical, numerical, statistical, Big Data, low-level/systems and scientific computing. I am also handy with optimisation, machine learning and AI, algorithm development, and theoretical issues.
• 🤔 Exploring: Currently, quantum computer science, LLMs and multi-modal machine learning models. In addition, I created open-source software aimed at assisting with mathematics writing and education. Finally, I am playing around with embedded systems programming and making chaotic microcontroller systems and machines!
• 💬 Ask me about: You can talk to me about most things, but I love discussing (and debating 😅) mathematics, computer science, physics, statistics and current trends within any of them. I also have a great deal of knowledge in the lore of Warcraft, Star Wars and Warhammer 40K, if you are keen to discuss!
• 📫 How to reach me: Best would be by email or LinkedIn. My eDoor is always open!
• Achievements: Besides my distinctions from my previous degrees, my favourite career achievement is deriving an alternate form of the Kalman filter equations logically identical to its original thing!

Much of my work is centred around high-performance and scientific computing topics, especially in Big Data regimes, currently within radio interferometric calibration. I come from a pure mathematics and theoretical computer science background that has allowed me to optimise and solve large computation problems from both mathematical and computational fronts.

I am highly interested in the discovery, design and optimisation of algorithms, particularly in the context of:

• (Bayesian) Statistical Estimation Algorithms
• Self-Improving Models and Techniques
• Machine Learning Models
• Quantum Computing

Most of my work is centred around research and development. Still, the majority of it requires me to continuously learn and master new techniques and technologies to improve computing, particularly in science.

Below is a list of my most notable projects thus far:

• Kalman Filtering and Smoothing in Calibration (KalCal): A calibration framework that uses the Kalman filter and smoother, alongside parameter-tuning techniques, to design and develop sophisticated radio interferometric calibration algorithms. The programming involved is distributed across several repositories.
• On-line Radio Interferometric Calibration Suite (ORCa): A new real-time or on-line calibration software consisting of the optimised computation server ORCa-Engine (Rust) and the web user-interface ORCa-UI (Python). The goal is to create a groundbreaking real-time calibration system using KalCal to improve current calibration strategies. This will be the start of creating a world-first radio interferometer preview system. It also tests the introduction and use of Rust within the Python, and C++ dominated radio astronomy software community.
• matque: A novel open-source software side-project that will generate mathematical questions and solutions for exercise, test or exam documents for various applications, e.g. homework, studying or formal tests. The goal is to make it highly customisable and diverse with natural language processing models while ensuring that solutions are 100% accurate (via a symbolic validation engine), reproducible and educationally insightful.
• wolf-whisper: An experimental open-source software that explores the ideas of creating software that translates human speech into LaTeX mathematics. It starts with OpenAI's whisper model for speech-to-text functionality, and it will end with a simple symbolic-to-tex using the sympy library. The current challenge is handling the middle layer that converts text-to-symbolic, e.g. "integral of x squared plus one" is integrate("x^2 + 1", "x"). An attempt was made via the Wolfram-Alpha API (hence the name), but the project has shifted towards exploring fine-tuned LLMs for this layer.

Research and Development Problem-Solving Public Speaking Teamwork Presenting Optimisation Big Data Technologies Mathematical Modelling Applied Research Mathematical Physics Algorithms Organisation Skills Statistical Modelling Statistical Analysis Machine Learning Data Visualisation Data Analysis Control Engineering Bayesian Statistics Bayesian Optimisation Automation Communication and Collaboration Object-Oriented Programming Functional Programming Software Development Software Design Debugging Virtualisation Databases Information Security Virtualisation

Rust Linux Python Shell Scripting LaTeX Docker Ubuntu/Debian Unix C/C++ Virtualisation

• I love video games!
• I played rugby at a high school provincial level.
• I have a level 4 Royal Schools of Music Certification in Piano.
• My favourite Warhammer 40K faction is the T'au!

Thank you for visiting my profile! Feel free to reach out if you're interested in collaborating or want to chat.