About me

With over five years dedicated to the ever-evolving field of machine learning, I've had the chance to contribute to transformative projects across various sectors. My journey has seen me seamlessly transition from the analytical depths of Data Science to the practical challenges of Machine Learning Engineering. Known for my proactive problem-solving approach and eagerness to tackle new challenges, my career is marked by engaging with the cutting edge of AI innovation. I've delved into areas like Large Language Models and Generative AI, driven by a passion to push the boundaries of what's possible in the ever-advancing realm of machine learning. For me, this field is more than a profession—it's an exciting pursuit to explore new frontiers.

Expertise In

  • Natural Language Processing

    Natural Language Processing

    In the wonderful world of NLP, I’ve had my fair share of adventures. Picture this: crafting custom named entities with SpaCy—like giving words their own unique identities. Then there's the detective work, using vector embeddings to unearth hidden patterns in search data for marketing teams. And let's not forget my foray into building foundational models for question-answering systems and text classification. It's like being a linguistic superhero, one algorithm at a time!

  • Machine Learning Development

    Machine Learning Development

    I've dabbled in the art of both supervised and unsupervised machine learning models, tackling everything from classification tasks to clustering. Imagine feature engineering and feature selection as my trusty toolkit, along with mastering the tricky balance of class imbalance. To top it off, I’ve used Docker and Kubernetes to deploy these models at scale, making sure they run smoothly in the big, wide world. It's like being a tech wizard, conjuring up solutions with a few lines of code!

  • Generative AI

    Generative AI

    Recently, I've ventured into creating LLM-based projects with some exciting outcomes. One project focuses on detecting vulnerabilities in smart contracts, while another involves strategy creation for sports. I've utilized LangChain to work with a variety of models, from open-source ones like Llama to closed-source powerhouses like OpenAI's GPT-4. It's like having a toolkit filled with the best AI models, ready to tackle any challenge!

  • Computer Vision

    Computer Vision

    In the exciting world of multimodal models, I developed a ViLBERT-based system to detect hateful content within memes. By leveraging the PyTorch framework, I created a model capable of understanding both visual and textual data. It’s been a fascinating journey blending image and text analysis to ensure memes remain fun and friendly!

Affiliations