Anastasios (Tasos) Stamoulakatos

Data Scientist & Applied AI Researcher

I am a Data Scientist at Satalia and hold a PhD in Applied AI. I specialize in bridging the gap between state-of-the-art academic research and real-world industrial systems—taking complex machine learning models from conceptual development to robust production pipelines.

My current focus lies in designing autonomous AI agent architectures, optimizing retrieval pipelines (RAG), and developing reliable machine learning applications.

🚀 Core Specialities

LLM & Agents

AI Agents & Tool Integration

Researching how LLM agents use environments and call tools (like web search, code execution, and database APIs) to complete multi-step tasks. Designing frameworks for agent evaluation, safety, and routing.

Retrieval

RAG & Web Search

Building high-performing Retrieval-Augmented Generation (RAG) systems that integrate real-time search capabilities and semantic indexers (like knowledge graphs and vector DBs) to ground LLM generations.

Computer Vision

Subsea Deep Learning

Completed PhD research on the automation of subsea pipeline survey video annotations. Developed 2D/3D CNNs and domain adaptation models to analyze high-resolution underwater pipeline footage.

Optimization

Industrial Data Science

Developing optimization algorithms and predictive models at Satalia. Previously worked at CMAC correlating particle microscopy with bulk powder properties for medicine manufacturing.

📰 Recent Satalia Research Lab Publications

Explore my latest technical articles on Large Language Models, Agentic Frameworks, and Retrieval-Augmented Generation:

Building Effective AI Agents

A brief guide and developer best practices for building robust agentic loops, handling tool usage, and managing planning modules in production.

2025 Read Article

Agentic Frameworks: A Comparison

An in-depth architectural comparison between modern multi-agent development frameworks like CrewAI, AutoGen, and LangGraph.