Hi!
I’m Igor
I’m Igor
I specialize in artificial intelligence (AI), computational linguistics and natural language processing (NLP), my expertise lies at the intersection of Ph.D.-level research, cutting-edge LLM architecture, and solving real-world business challenges.
From developing intelligent assistants (including systems that learn and answer using your company's private documents) to fine-tuning complex AI architectures, I use this unique blend
of skills to push the boundaries of what is possible with AI and deliver tangible results for business.
From developing intelligent assistants (including systems that learn and answer using your company's private documents) to fine-tuning complex AI architectures, I use this unique blend
of skills to push the boundaries of what is possible with AI and deliver tangible results for business.
Generic AI is a commodity.
Your Own AI is a competitive moat.
Your Own AI is a competitive moat.
My Research Journey
PhD in Information and Communication Technologies
at Pompeu Fabra University
at Pompeu Fabra University
Research Field: Dialog Understanding
I am developing AI solutions designed to enhance human-machine synergy. My doctoral thesis is dedicated to the study of dialogue understanding in LLMs; I work on uncovering the "black box" behavior of these models to analyze how they interpret interaction patterns and complex conversational structures.
I am developing AI solutions designed to enhance human-machine synergy. My doctoral thesis is dedicated to the study of dialogue understanding in LLMs; I work on uncovering the "black box" behavior of these models to analyze how they interpret interaction patterns and complex conversational structures.
Master in Intelligent Interactive Systems at Pompeu Fabra University
Research Field: Dialog Systems
Foundational training in computational linguistics, natural language processing, and the design of adaptive, user-centric intelligent systems. This provides the core understanding necessary for effective dialogue management and RAG architecture.
Foundational training in computational linguistics, natural language processing, and the design of adaptive, user-centric intelligent systems. This provides the core understanding necessary for effective dialogue management and RAG architecture.
Certified C/C++ Programmer
Intensive training in core software design, application programming, and foundational C/C++ logic. This rare skill set ensures all custom AI solutions are robust, highly optimized, and scalable.
I actively contribute to AI and NLP research. You can find my latest papers and projects here
Publications
I specialize in AI and NLP development, focusing on LLM architectures, natural language understanding, and AI-driven automation. My expertise includes deep learning, chatbot development, and AI model fine-tuning using modern frameworks and cloud solutions.
Technical Expertise
GPTs, LLaMAs, Mistral, Qwen, Gemma, Mamba, NLLB, Molmo, MiniCPM, BERT family and many others
Python, C/C++, SQL, Golang, Matlab
Docker, AWS, Google Cloud
PyTorch, Transformers, PEFT, TRL,TransformerLens, SAELens, VLLM, LangChain, LLamaIndex, OpenAI API
NLTK, SpaCy, RASA, Gensim, Natasha
Professional Timeline
Research Engineer at Barcelona Supercomputing Center
2024-present
2024-present
LLM development automatization: Developing
state-of-the-art automated pipelines for LLM deployment and evaluation, including tooling for LLM quantization.
state-of-the-art automated pipelines for LLM deployment and evaluation, including tooling for LLM quantization.
NLP Engineer — Custom Solutions
& Consulting)
2022–Present
& Consulting)
2022–Present
Designing and implementing end-to-end RAG systems (e.g., PreggApp), constructing proactive assistants (GPT/Gemini), and building AI Agents for UI-testing automation (ScreenMate AI).
LLM Engineer at KVINT
2024
2024
Developing an LLM-based Voice Assistant and transferring complex rule-based pipelines to robust LLM architectures.
C/C++ Programmer at CSoft Company
2022–2024
2022–2024
Core development and support for CAD systems and 3D engines, focusing on performance optimization and shaders programming (DX11/DX12, Vulcan).