Open to opportunities

Georgios Traskas

@georgiostraskas

Message

Senior AI & ML engineer building production-grade LLM systems, RAG pipelines, and observable Python backends.

Greece

Message

What I'm looking for

I’m looking for a team where I can build and ship production LLM systems and RAG pipelines with strong security guardrails, structured outputs, and observability—using Python backends and AWS to turn model performance into real business impact.

I’m a Senior AI and ML Engineer with 8+ years of experience building production-grade LLM systems, RAG pipelines, and data science solutions. I’ve delivered ML-driven predictive models and production observability infrastructure, reporting directly to the Head of Science and Innovation and to the Founder and CEO.

I design end-to-end pipelines—from AWS data workflows and feature engineering to reliable deployment and monitoring—so teams can move faster with confidence. I’ve also completed 90+ global consulting engagements, delivering deep learning models, automated pipelines, and AI-powered applications across healthcare, energy, and sports science.

My work is centered on production robustness: multi-layer security guardrails, vector retrieval architectures, and structured LLM output contracts. I build hybrid retrieval (BM25 + vector search), orchestration with LangChain, and analytics/reporting that translate model performance into actionable business decisions.

Experience

Work history, roles, and key accomplishments

Current

Independent AI Consultant

Current

Data Science Solutions

Jul 2020 - Present (5 years 11 months)

Delivered 90+ global AI/ML engagements via Upwork with a 100% job success score across healthcare, energy, and retail. Built and optimized production ML solutions including a 97% accurate deep-learning classification model for embedded deployment (TFLite quantization) and a 95% accurate clinical decision-support model using SMOTE and cross-validation.

Machine Learning Deep Learning Python TFLite Quantization SMOTE Cross Validation Classification Statistical Modeling

Data Scientist & ML Engineer

Predictive Fitness, Inc.

May 2022 - May 2026 (4 years)

Developed and deployed production ML pipelines for athlete fitness telemetry, including time-series forecasting and workout outcome prediction. Built AWS-based automated data workflows and redesigned alerting with Grafana and Sentry, reducing false alert noise by 90%+, while improving trial conversion modeling via causal-feature logistic regression.

Python AWS Lambda Athena Time Series Models GraphQL Grafana Sentry

Chemist & Automation Engineer

Centre for Research & Technology Hellas

Dec 2007 - Jul 2022 (14 years 7 months)

Automated analytical data pipelines, increasing system throughput by 80% and reducing delivery times by 50%. Applied statistical analysis and experimental design, producing reporting tools used by scientific teams and management.

Automation Systems Data Pipelines Throughput Statistical Analysis Experimental Design Reporting Tools Data Processing