Skip to content

Software Engineering & AI

Building systems and automation workflows that turn complex concepts into usable applications.

About Me

I'm a Computer Science student at Saarland University with a strong interest in software engineering and applied AI.

My projects span the full stack: from low-level C and assembly work to backend services, cloud deployment, and user-facing applications. I have built a C-based SAT and Sudoku solver platform, a MIPS Assembly IBAN validator exposed through an API, and applications using FastAPI, Redis, Docker, and AWS.

I enjoy building software end to end: understanding the core algorithm, designing the system around it, deploying it reliably, and making the final result usable through clean demos and interfaces. I am also actively exploring AI and automation tools such as LLM APIs and n8n in practical backend workflows.


Featured Projects

C SAT & Sudoku Solver Platform

Backend Systems

Engineered a high-performance Sudoku and constraint solving platform combining a low-level SAT solver (C) with distributed async job processing (FastAPI, Redis).

  • Designed a FastAPI backend following REST best practices.
  • Implemented an asynchronous job processing pipeline using Redis and background workers.
  • Set up CI/CD via GitHub Actions and implemented API & unit tests.

Tech: C, FastAPI, Redis, Docker, Github Actions, AWS(EC2, RDS, S3)

Live Demo Sudoku Solver | Live Demo Distributed Sudoku Solver
Live Demo SAT Solver

MIPS IBAN Validator and Generator App

Systems + Backend

Full-stack IBAN <-> KNR/BLZ validator combining financial identifier processing with a modern FastAPI backend, featuring persistent history and privacy-conscious logging.

  • MIPS Assembly core exposed through a FastAPI backend with strict input validation through pydantic schemas.
  • Added SQLite-backed history, structured logging and privacy masking for sensitive financial identifiers before persistence.
  • Dockerized and deployed to AWS, automated CI/CD via Github Actions

Tech: MIPS, FastAPI, JavaScript, HTML, SQLite, Docker, Github Actions, AWS

Live MIPS IBAN Validator

AI Coding Agent with MCP Tool Integration

AI / Agents

Built a terminal-based AI coding assistant from scratch, then extended it with MCP servers exposing existing deployed projects — enabling the agent to autonomously call the Sudoku Solver and IBAN Validator APIs as tools.

  • Implemented the full agent loop: LLM decides action → tool executes → result fed back → iterate
  • Wrote MCP servers wrapping the Sudoku Solver and MIPS IBAN Validator REST APIs, exposing them as structured tools
  • Agent autonomously selects and calls the correct tool based on natural language task description

Tech: Python · Anthropic Claude API · MCP · Tool Calling · Agent Architecture · FastAPI

Multi-Agent Workflow with Google ADK

AI / Agents

Designed and built a multi-agent system using Google's Agent Development Kit (ADK), orchestrating specialised sub-agents to decompose and complete complex tasks autonomously.

  • Defined agent roles, tool bindings, and delegation logic using the ADK framework
  • Integrated external APIs and services as callable tools within the agent graph
  • Implemented session state and inter-agent communication for multi-step task completion

Tech: Python · Google ADK · Gemini API · Tool Calling · Multi-Agent Systems

Software Design Engineering Lab - Group and Individual Phase

Software Engineering

Multi-phase engineering lab focused on architecting and extending a complex simulation system under review and testing constraints.

  • Co-designed modular architecture using UML and defended design decisions
  • Implemented controllers, handlers, and event-driven extensions
  • Improved validation and testing strategy, contributing to a 98/100 pass rate

Tech: Kotlin, UML, Unit & Integration Testing

Project Details Individual Phase
Project Details Group Phase

Tech Stack

Languages

Python
JavaScript
Java
C
Kotlin
Scala
RISC V

Backend & APIs

FastAPI
Django REST
Redis
PostgreSQL
SQLite

Cloud & DevOps

AWS
Docker
GitHub Actions
GitLab CI
Vercel
DigitalOcean

Testing

Pytest
JUnit

Experience

Tutor — Mathematics for Computer Science

Universität des Saarlandes

  • Taught linear algebra and discrete mathematics
  • Focus on rigorous reasoning and proof techniques
  • Strengthened students' analytical and abstraction skills

Software Praktikum (Software Design Engineering Lab)

Universität des Saarlandes

  • Selective 7-week software engineering lab building a complex multi-agent simulation system
  • Co-designed modular architecture using UML class, sequence, and state diagrams; defended decisions in formal design and code reviews
  • Implemented core controller logic with new controllers, handlers, and an event-driven extension
  • Enhanced JSON parsing with additional schema rules, cross-validation, and robust error handling
  • Led the team's testing strategy, mentoring teammates on unit and integration testing, achieving 98/100 functional test pass rate in group and 48/50 in individual phase.

Contact