I'm an AI engineer with 3+ years building production audio/speech ML and LLM applications. Background includes 6 years of .NET/Azure backend, which occasionally pays off when ML meets enterprise infrastructure.
I started my career in .NET enterprise development — building CRM integrations, reporting systems, and Azure-hosted applications for clients in banking, automotive, and tourism.
Over time, I shifted toward machine learning. Today I work on RAG, LLM integration, transformer-based audio models, and ML classification tools.
Fine-tuning audio generation models and building RAG pipelines.
Audio ML pipelines — classification, generation. Transformer-based architectures for audio.
I'm Microsoft certified Data Scientist Associate. Production-grade APIs and cloud infrastructure. Strong .NET ecosystem background with Azure services.
Strong foundation in software development — clean code, algorithms, data structures, and object-oriented design.
Multi-speaker speech-to-speech pipeline with speaker embeddings. Designing experiments for Runpod with PyTorch.
Built a reporting SPA from scratch. Domain-driven design backend hosted on Azure.
Tourism prediction app backend. NetLogo simulation integration with university researchers.
Banking CRM integrations. Azure Functions, SSRS reports, MS Dynamics customization.
MS Dynamics CRM plugin development. Customer requirements analysis.
Attendance system in PHP/Nette. WinForms maintenance app in C#.
ML classification tools for audio engineers using transformer-based models. Architecture experiments to improve production accuracy.
Fine-tuning local LLMs for an educational app for older adults. Recommendation algorithms and React frontend.
4+ years building strategies with Freqtrade. GBDT models via FreqAI for crypto market prediction and automated decisions.
Conversational chatbot for a Czech e-shop recommending products using a local LLM with optimized prompt engineering.
Each point represents a speech sample from one of 7 European languages. Audio files are converted into 768-dimensional embeddings using a ContentVec transformer model, then projected into 3D space with UMAP. The resulting clusters show that the model captures language-specific acoustic patterns — similar languages appear closer together.
Open to Software engineering roles, freelance projects, and interesting collaborations.