Skip to content
View VasilisStavrianoudakis's full-sized avatar
2 followers · 1 following

Achievements

Achievement: Pull Shark

Achievements

Achievement: Pull Shark

Block or report VasilisStavrianoudakis

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
VasilisStavrianoudakis/README.md

Hi, I'm Vasilis 👋

AI & Data Manager based in Athens, Greece. I've spent most of my career building machine learning systems and leading the teams that ship them — NLP, LLMs, computer vision, OCR, ASR, forecasting — usually on the kind of messy, sensitive, real-world data that never looks like the benchmark.

Lately my focus has shifted to agentic AI. I help enterprises figure out where agents actually belong in their workflows, from the software development lifecycle to wider GenAI strategy, and how to get them into production without treating security and governance as an afterthought.

I've been leading AI/ML teams for a few years now — managing engineers, setting technical direction, working closely with product and clients — but I've made a point of staying hands-on. I will always like the AI work: building Agentic workflows, fine-tuning LLMs, taming imbalanced datasets, and making services run reliably.

If there's a thread through my work, it's turning open-ended problems into systems that hold up on real, messy data.

Get in touch

Pinned Loading

  1. Attribute-Conditioned-Facial-Image-Generation-Through-Contrastive-Learning Attribute-Conditioned-Facial-Image-Generation-Through-Contrastive-Learning Public

    MSc in Artificial Intelligence Thesis: A PyTorch multi-GPU, attribute-conditioned facial image generation framework using contrastive learning to improve image quality and attribute control.

  2. Contrastive-Learning-Based-Text-Classification Contrastive-Learning-Based-Text-Classification Public

    A Supercharged two-phase transformers based text classification training pipeline with: contrastive pre-training + asymmetric loss. Perfect for imbalanced datasets! 🚀

    Jupyter Notebook

  3. Deep-Learning-Image-Captioning Deep-Learning-Image-Captioning Public

    An end-to-end deep learning solution for automatic image captioning using convolutional neural networks (CNN) as encoders and recurrent neural networks (RNN) as decoders in PyTorch.

    Jupyter Notebook

  4. HelpSteer-ComplexityPrediction HelpSteer-ComplexityPrediction Public

    Predict the complexity of AI-generated responses using a fine-tuned transformer model trained on NVIDIA's HelpSteer dataset using an asymmetric quantile loss function for imbalanced data.

    Jupyter Notebook