SYSTEM PROFILE // ONLINE
ML/AI Engineer · Software Developer · M.Sc. Bioinformatics · B.Sc. Computer Science and Engineering
I deeply enjoy working on diverse cross-functional projects that challenge me, expand my skills, and allow me to build meaningful technical solutions.
LOADOUT // TECH
ARCHIVE // 001
Implementation of Inverse Reinforcement Learning algorithms with Python solving the Mountaincar-v0 experiment.
Demo-Project: End-to-end lung cancer risk prediction system with secure doctor authentication, encrypted patient data, and decision-tree predictions.
HeartPredict is a Python library designed to analyze and predict heart failure outcomes using patient data.
This is a demo project showing how to create an MCP client with a local LLM in both Python and TypeScript. It also demonstrates how to configure an MCP server in Python (with FastAPI) and in TypeScript.
This is my individual project for the module Research Software Engineering in SS24. The task was to analyze a dataset from genesis.destatis using Python and to find interesting aspects and potential questions that could be explored using this data.
CoAler (Core Aligner), a tool capable of efficiently calculating multi-alignments for dozens of molecules from scratch within minutes on a standard desktop computer.
OntoGraph is a C++ tool that converts RDF/XML ontologies into graph structures, enabling traversal and search using standard graph algorithms. It is designed for researchers and developers working with ontologies who need to analyze, traverse, or manipulate ontology data as graphs. Note: CLI is under construction.
Simulation of a bee colony, where the bees try to discover all food sources in the simulated world. The Bee Colony Optimization algorithm is used as core algorithm.
L.O.F.I is a lofi-themed educational and productivity app that helps you stay focused. Features: Pomodoro Timer, Statistics, Themes, Achievements and To-Do List.
The ByakuganVisualizer is a Python tool for image comparison and highlighting differences, facilitating tasks such as testing and quality assurance, with an added color filter feature for correcting images for color-blind users.
This tool is created to convert MP4 videos into ASCII animations.
DOSSIER // CV
RECORDS // CERTS
SPECIALIZATION
A Google Cloud specialization focused on production ML systems, natural language processing, computer vision, and recommendation systems.
Google Cloud · 2026 · This course explains how to build high-performing machine learning systems for production. It covers training and inference strategies, distributed TensorFlow, TPUs, and key practices for creating reliable ML systems that go beyond simply making accurate predictions.
Google Cloud · 2026 · This course introduces the products and solutions to solve NLP problems on Google Cloud. Additionally, it explores the processes, techniques, and tools to develop an NLP project with neural networks by using Vertex AI and TensorFlow.
Google Cloud · 2026 · This course introduces different computer vision use cases and machine learning strategies, ranging from pre-built ML APIs and AutoML Vision to custom image classifiers using linear models, DNNs, and CNNs.
Google Cloud · 2026 · In this course, you apply your knowledge of classification models and embeddings to build a ML pipeline that functions as a recommendation engine.
SPECIALIZATION
This IBM Professional Certificate develops advanced generative AI engineering skills through hands-on work with prompt engineering, RAG, multimodal AI, autonomous multi-agent systems, and frameworks such as LangChain, LangGraph, CrewAI, AG2, BeeAI, and MCP.
IBM · 2026 · This course introduces the fundamentals of generative AI application development and teaches you how to build full-stack GenAI applications using prompt engineering, LangChain, agents, Flask, Gradio, structured JSON outputs, and different language models.
IBM · 2026 · This course introduces the fundamentals of Retrieval-Augmented Generation (RAG) and teaches you how to build interactive RAG applications using LangChain, LlamaIndex, Gradio, embeddings, and vector databases.
IBM · 2026 · This course introduces vector databases for Retrieval-Augmented Generation (RAG) and teaches you how to use ChromaDB, embeddings, vector operations, and similarity search to build recommendation and retrieval systems.
IBM · 2026 · This course covers advanced Retrieval-Augmented Generation (RAG) techniques and teaches you how to build high-performance RAG applications using FAISS, ChromaDB, LangChain, LlamaIndex, advanced retrievers, and Gradio.
IBM · 2026 · This course introduces multimodal generative AI and teaches you how to build applications that combine text, speech, images, and video using models and tools such as Whisper, DALL·E, Sora, Llama, Flask, Gradio, and multimodal retrieval techniques.
IBM · 2026 · This course introduces the fundamentals of AI agents and teaches you how to build goal-oriented agentic systems using LangChain, tool calling, LCEL, custom tools, memory, data analysis agents, and SQL agents.
IBM · 2026 · This course teaches you how to build advanced agentic AI systems with LangChain and LangGraph, including stateful workflows, memory, Reflection, Reflexion, ReAct, multi-agent orchestration, and agentic RAG with dynamic query routing.
IBM · 2026 · This course teaches you how to design and build scalable multi-agent AI systems using LangGraph, CrewAI, BeeAI, and AutoGen, with a focus on workflow patterns, orchestration, tool use, memory, and agent collaboration.
IBM · 2026 · This course introduces the Model Context Protocol (MCP) and teaches you how to build secure AI agents by creating MCP servers and clients, connecting tools and data sources, managing transports and permissions, and designing controlled multi-agent workflows.
IBM · 2026 · This capstone course brings together RAG and agentic AI concepts by having you build an end-to-end multimodal AI system with structured data, embeddings, vector search, multi-agent workflows, MCP integration, evaluation, and a Gradio interface.
COURSES
Additional completed Google Cloud coursework focused on MLOps and production ML infrastructure.
Google Cloud · 2026 · This course introduces participants to MLOps tools and best practices for deploying, evaluating, monitoring and operating production ML systems on Google Cloud. MLOps is a discipline focused on the deployment, testing, monitoring, and automation of ML systems in production.
LOGBOOK // POSTS
ByakuganVisualizer started from a personal motivation: I have a mixed form of deuteranomaly and protanomaly, which made me interested in how color perception can be simulated, analyzed, and improved through image processing. While researching the topic, I found the paper: An Adaptive Fuzzy-Based System to Simulate, Quantify and Compensate Color Blindness
The paper describes methods for simulating and compensating color blindness using fuzzy parameters, linear transformations, and histogram-based operations.
The derived transformation matrix became the foundation for the color correction feature in
ByakuganVisualizer. With my personal settings deuteranomaly = 2 and protanomaly =
2, the tool transforms color-blind test images in a way that makes the hidden numbers
distinguishable for me.
Example: before correction and after correction . After applying the correction, I am able to see every number in the test image. You can view the full project on GitHub or feel free to test the live demo on Hugging Face .
PERSONAL // SIDE DATA
I enjoy listening to music while working and studying, which inspired me to start generating my own beats. I create atmospheric, lofi, and trap-inspired music.
Visit my YouTube channel