Hackathon Portal
AI Tinkerers - Budapest
Team

Lab5

This team is at maximum capacity.

Project Concept

No description has been added yet.

Entry

Status: Submitted

Last saved: December 11 at 9:31 PM CET

Team Roster (team is at max capacity)

Message board not available for this team yet.

Munkhchimeg Sergelen Team Lead RSVP Approved

Software Developer at Lightbloom AI Lab
Voice pipeline engineer and team lead. Implemented complete voice system using ElevenLabs APIs: microphone capture (mic.js), Speech-to-Text transcription (stt.js using ElevenLabs Scribe), Text-to-Speech synthesis (tts.js with custom cloned voice). Built OpenAI GPT-4o-mini orchestrator for conversational AI. Debugged STT/TTS integration issues and ensured end-to-end voice flow.
I’m a software developer with a background in electrical engineering. I’ve worked with multilingual ASR systems and large language models, and I’m particularly focused on building AI-powered accessibility technologies. My recent work involves experimenting with speech recognition for low-resource languages and developing prototypes that translate spoken language into avatar-based sign language.
AI for accessibility, assistive tech, speech processing, low-resource languages, and real-time ASR→LLM→3D pipelines. I enjoy collaborating with developers and designers on creative AI, interactive tools, and avatar-based communication systems.
SignMate: a speech-to-sign avatar prototype using ASR, LLMs, and 3D animations. ASR Evaluation: benchmarking multilingual speech models on Mongolian, Hungarian, Spanish, and French. 3D Avatar Work: experimenting with Ready Player Me and Mixamo rigs for custom motion generation.

Muhammad Eman Aftab RSVP Approved

Software Developer at Lightbloom
Menu reader and cash detector modules. Developed Python scripts (menu_reader.py, cash_detector.py, smart_detector.py) using OpenAI Vision API to analyze food menus (extracting items, prices, categories) and detect/count cash (bills, coins, total amount). Created natural language text formatting for TTS output.

Kevin Nava RSVP Approved

Computer Vision Engineer at Lightbloom
Backend development and image analysis pipeline. Built the Python backend (backend/model.py, backend/open_api.py) for processing camera frames. Integrated OpenAI Vision API for scene analysis. Merged multiple branches and handled backend-to-frontend data flow.
+6 as software developer , Msc. Image Processing and computer Vision
AI
3D Reconstruction

Zeineb Turki RSVP Approved

software developer at Lightbloom
Frontend UI and final integration. Built React-based accessible UI with mobile support (Capacitor for iOS/Android). Implemented camera preview, conversation interface, and status displays. Completed final integration merging all team branches (voice-pipeline, navigation-reader-helper, kevin) into working prototype. Added comprehensive setup documentation.
I’m Zeineb Turki, a computer science engineering student who loves building things, solving problems, and learning across disciplines. I enjoy exploring how software, systems, and emerging technologies can shape the future. I’ve worked on projects ranging from networking and cybersecurity to game development and automation, and I thrive in environments where I can experiment, collaborate, and turn ideas into real prototypes. I’m driven by curiosity and the desire to create meaningful, well-designed solutions.
Software engineering, systems design, networking, automation, robotics, creative coding, and emerging technologies. I’m especially drawn to projects that combine engineering with real-world impact, whether in aerospace, AI, security, health care, or interactive systems. I love learning new tools, experimenting with bold ideas, and exploring cross-disciplinary tech.
Currently developing AutoScan AI, an early-detection concept that analyzes MRI scans using multimodal techniques to identify tumor patterns at a much earlier stage. Exploring data preprocessing, image-analysis pipelines, and model-evaluation strategies to build a fast, reliable medical-imaging assistant. Also experimenting with aerospace–AI crossover ideas and practical bioinformatics tools.