Lazaros Koutsianos
Machine Learning Engineer & Data Scientist with a focus on Retrieval Augmented Generation
Experience
RAG Webinar: Deep Dive and Use Cases
SHI GmbH
- Design, preparation and delivery of a webinar on 'RAG in Practice: How publishers create real value with AI'
- Preparing technical and strategic content on Retrieval Augmented Generation (RAG) for a mixed audience from the publishing industry
- Presenting specific use cases, technical backgrounds, common challenges and solution approaches when using RAG
- Providing practical insights into data preparation, model selection and output optimization in the context of digital publishing portals
- Conceptual and technical preparation of the webinar
- Selecting and presenting practical use cases from the publishing environment
- Developing technical backgrounds for implementing RAG systems
- Presenting and explaining typical challenges and solution strategies
- Large Language Models (LLMs)
- Retrieval Augmented Generation (RAG)
POC delivery and concept development for a RAG-based chatbot
SHI GmbH
- Creating a POC where customers can interactively evaluate parameters for a RAG-based chatbot
- Creating a concept document for implementing a RAG-based chatbot
- Technical responsibility for selecting RAG components (chunking, retrieval, embedding model, LLM) and prototyping in Python
- Vectorizing and storing data in a vector database
- Creating and optimizing the system prompt using prompt engineering
- Testing and deploying local LLMs
- Creating a concept document with all technical requirements and infrastructure components for chatbot implementation – both with open source and proprietary models
- ML-Ops
- Python
- Pytorch
- LLMs
- Hugging Face
- Ollama / vLLM / llama.cpp
- Langchain
- LlamaIndex
- Chroma, Elasticsearch, Apache Solr
- SentenceTransformers
- OpenAI Embeddings
- Hugging Face Transformers
- Flask
Author / Co-author: White paper on Retrieval Augmented Generation
SHI GmbH
- Researching and writing a white paper on Retrieval Augmented Generation as the author for an internal project
- Co-authoring another white paper in cooperation with the German Publishers and Booksellers Association
- Presenting technology basics, use cases, as well as opportunities and challenges of RAG in a professional information environment
- Researching, structuring and developing the content of the texts
- Clearly presenting technical concepts for a mixed professional audience
- Coordinating with co-authors and aligning the content for joint publication
- Bringing in practical knowledge from previous projects to illustrate use cases
- Large Language Models (LLMs)
- Retrieval Augmented Generation (RAG)
RAG-based chatbot
SHI GmbH
- Implementing a chatbot based on the principle of Retrieval Augmented Generation
- Answers based solely on provided documents
- Technical responsibility for selecting RAG components (chunking, retrieval, embedding model, LLM) and prototyping in Python
- Vectorizing and storing data in a vector database
- Creating and optimizing the system prompt using prompt engineering
- Evaluating the model responses and iteratively improving the result quality
- Testing and deploying local LLMs
- Supporting the practical implementation of the chatbot
- ML-Ops
- Python
- Pytorch
- LLMs
- Hugging Face
- Ollama / vLLM / llama.cpp
- Langchain
- LlamaIndex
- Chroma, Elasticsearch, Apache Solr
- SentenceTransformers
- OpenAI Embeddings
- Hugging Face Transformers
- Flask
Log file analysis
SHI GmbH
- Building an analysis system with Elasticsearch, Logstash and Kibana to improve search functionality in the online shop
- Using Apache Solr logs and web server logs as analysis basis
- Configuring Solr to log the required data at INFO level
- Installing Elasticsearch, Logstash and Kibana
- Configuring Logstash including defining suitable Grok patterns for indexing Solr logs
- Defining an ingest pipeline in Elasticsearch to determine geo information based on provided IP addresses
- Analyzing sample data for Solr and web logs and creating appropriate mappings in Elasticsearch as well as data views in Kibana
- Creating and assembling visualizations on dashboards in Kibana
- Apache Solr
- Elasticsearch
- Logstash
- Kibana
Member of Search & Analytics Team
SHI GmbH
Dense Vector Search
SHI GmbH
- Comparison of traditional text-based search results with vector search results in Apache Solr 9
- Implementation and documentation of a demo version for the "Dense Vector Search" feature
- Evaluation of suitable machine learning models from Hugging Face to generate vectors for vector search
- Implementation of methods to create vector representations of test data
- Configuration of schema and request handlers and formulation of search queries in Apache Solr for using Dense Vector Search
- Apache Solr
- Python
- PyTorch
- Pandas
- pysolr
Machine Learning-based Anomaly Detection in Production Processes
SHI GmbH
- Design and implementation of a machine learning model for automated detection of anomalies in production data
- Identification of quality deviations based on process data
- Advising on data-driven optimization of manufacturing processes
- Analysis and preprocessing of sensor data from production equipment (temperature, pressure, vibration, cycle time)
- Development and comparison of anomaly detection models (Isolation Forest, One-Class SVM)
- Validation of detected anomalies using historical quality and scrap data
- Visualization of identified process deviations over time and presentation of results
- Python
- scikit-learn
- Pandas
- NumPy
- Seaborn
- SQL
Data Integration
SHI GmbH
- Improving and consolidating financial data
- Integrating SAP data into CCH Tagetik
- Creating concept documents
- CCH Tagetik
NLP-based Classification of Service and Error Reports
SHI GmbH
- Development of an NLP-based model for automatic classification of technical incident reports and error descriptions from a maintenance management system
- Identification of recurring error types to support maintenance strategy
- Improving the analysis of unstructured data
- Text preprocessing (tokenization, lemmatization, stopword filtering)
- Feature extraction using TF-IDF and n-grams
- Training a text classification model for error classification
- Model comparison (Logistic Regression, SVM, Naive Bayes)
- Integration of the model into an internal monitoring system
- Python
- scikit-learn
- spaCy
- NLTK
- Pandas
- Matplotlib
Web Development & Linguistics
SHI GmbH
- Localization and optimization of software for users of different languages
- Optimization of PHP code and creation of HTML web pages for software and the online shop
- Independent development of HTML web pages
- Preparation of translations for the language course content and the online shop
- Voice actor for language courses, voicing all Greek translations
- PHP
- HTML
Industries Experience
See where this freelancer has spent most of their professional time. Longer bars indicate deeper hands-on experience, while shorter ones reflect targeted or project-based work.
Experienced in Information Technology (7.5 years), Education (5.5 years), and Manufacturing (2.5 years).
Business Areas Experience
The graph below provides a cumulative view of the freelancer's experience across multiple business areas, calculated from completed and active engagements. It highlights the areas where the freelancer has most frequently contributed to planning, execution, and delivery of business outcomes.
Experienced in Information Technology (7.5 years), Product Development (7 years), Business Intelligence (5.5 years), Operations (2.5 years), Quality Assurance (2.5 years), and Research and Development (1 year).
Summary
Lazaros has a strong analytical mindset and the ability to quickly dive into new topics and technologies. He brings analytical thinking, quick comprehension, and a high level of initiative. Through his academic and professional career, he has learned to look at tasks from different angles and critically question existing processes. He works efficiently both in teams and independently and keeps an overview even in new or challenging situations.
Skills
Information Retrieval & Retrieval Augmented Generation
Large Language Models & Natural Language Processing
Machine Learning & Deep Learning
Data Engineering & Ml-ops
Semantic & Hybrid Search In Vector Databases
Search & Big Data Technologies, Especially Apache Solr And Elasticsearch
Translating Business Requirements Into Technical Solutions
Python
Html
C++
Sql
Unix Command Line Tools
Shell Scripts
Latex
Windows
Linux
Macos
Apache Solr
Elasticsearch
Langchain
Llamaindex
Pandas
Numpy
Spacy
Apache Nifi
Pytorch
Logstash
Embeddings
Llmops
Mlops
Ollama
Prompt Engineering
Prompt Injection
Vector Stores
Vllm
Git
Googleads
Jira
Cpm Software Cch Tagetik
Pysolr
Kibana
Ci/cd
Cuda (Nvidia)
Docker
Gitops
Jenkins
Mysql
Pycharm
Visual Studio Code
Overleaf
Texmaker
Implementation Of Retrieval Augmented Generation Systems
Development And Integration Of Retrieval Infrastructures
Improving And Optimizing Search-driven Applications
Dataflow Management, Data Processing & Analysis
Application Of Machine Learning Methods (Clustering, Classification, Recommendations)
Natural Language Processing
Management-level Documentation & Presentations
Training Sessions On Artificial Intelligence
Keynote Speaker At Several Public Events
Languages
Education
Ludwig-Maximilians-Universität München
Bachelor of Science · Computational Linguistics · Munich, Germany
Ludwig-Maximilians-Universität München
Bachelor of Arts · History and Archaeology · Munich, Germany
Certifications & licenses
Elastic Certified Engineer
Profile
Frequently asked questions
Do you have questions? Here you can find further information.
Where is Lazaros based?
What languages does Lazaros speak?
How many years of experience does Lazaros have?
What roles would Lazaros be best suited for?
What is Lazaros's latest experience?
What companies has Lazaros worked for in recent years?
Which industries is Lazaros most experienced in?
Which business areas is Lazaros most experienced in?
Which industries has Lazaros worked in recently?
Which business areas has Lazaros worked in recently?
What is Lazaros's education?
Does Lazaros have any certificates?
What is the availability of Lazaros?
What is the rate of Lazaros?
How to hire Lazaros?
Average rates for similar positions
Rates are based on recent contracts and do not include FRATCH margin.
Similar Freelancers
Discover other experts with similar qualifications and experience
Experts recently working on similar projects
Freelancers with hands-on experience in comparable project as a RAG Webinar: Deep Dive and Use Cases
Nearby freelancers
Professionals working in or nearby Augsburg, Germany