Machine Learning Engineer focused on building AI systems that move from raw data to production applications.
My work centers around LLM-powered applications, retrieval systems, and perception pipelines, combining machine learning models with reliable backend infrastructure.
Currently exploring real-time AI assistants, RAG pipelines, and robotics perception systems.
Languages
Python · C/C++ · SQL · Shell
ML / AI
PyTorch · Transformers · Whisper · Pyannote · Sentence Transformers
LLM Systems
LangChain · ChromaDB · Vector Search · RAG
Robotics / Vision
ROS2 · Gazebo · OpenCV
Infrastructure
FastAPI · Docker · SQLAlchemy · NiceGUI
Speech → knowledge pipeline that converts meetings into searchable information.
Features
- speech transcription (Whisper)
- speaker diarization
- semantic search over conversations
- retrieval-augmented QA with LLMs
Stack
Whisper · Pyannote · Sentence Transformers · ChromaDB · LangChain · Docker
👉 https://github.com/SIMETRAL55/MeetBot
Simulation environment for experimenting with multi-view perception and spatial inference using thermal cameras.
The system reconstructs a target’s 3D position from multiple camera observations, mimicking a motion-capture setup used in robotics and intelligent sensing systems.
Features
- multi-camera image processing pipeline
- 3D localization via geometric triangulation
- sensor data fusion from multiple viewpoints
- real-time visualization in RViz
Stack
ROS2 · Gazebo · OpenCV · Python
👉 https://github.com/SIMETRAL55/Gazebo-Motion-Capture




