Michał Żarnecki Portfolio

Lecture: Retrieval Augmented Generation – replace complex logic with a text model
Abstract: Have you ever encountered code with so many conditions and processing scenarios that it was almost impossible to maintain and extend it? What if we replaced it with an automatically generated, self-improving algorithm? In recent years, machine learning as a field of artificial intelligence has become an effective tool for creating systems and applications. The dynamic development of models based on artificial neural networks means that the programming of complex business rules and services based on the summarization and extraction of information can be successfully replaced with machine learning models. In this presentation, you will see a case study illustrating the process of building a simple application based on RAG (Retrieval Augmented Generation) in PHP using a large text model (LLM) to effectively find precise answers in a database of unstructured text data.

https://chat.companyhouse.de/

AI-based chatbot for retrieving reliable, up to date and precise information about companies.
Chatbot is based on streamlit framework and uses vector database based on postgres pg_vector extension to store and access trade register documents.
Application is using large language model (LLM) Llama3 together with retrieval augmented generation (RAG) approach which allows to ask and get response to any question related company and managers history as well as financial condition and important changes.
Together with response also source documents are listed making this approach reliable business intelligence tool.

Responsibilities:

• build application prototype
• implement application code parts
• implement authentication mechanism
• specify and coordinate works related to building chatbot interactions
• specify and coordinate works related to sychronizing in real time source documents and make them accessible for LLM
• measure answers quality

Course scenario:

1. Definition and applications of machine learning
   • Data deluge and the definition of machine learning
   • Machine learning examples and related fields of knowledge
   • Types of machine learning
2. Machine learning tools used in the course
   • Programs used in the course
   • Orange Data Mining
   • Jupyter Lab
3. Supervised machine learning
   • Machine learning process
   • Data collection, labeling and analysis
   • Feature engineering and division into training and testing sets
   • Model training and evaluation
   • Model export, corrective actions
   • Regression example
   • Classification example

ML with PHP – replace complex business logic with machine learning models

Abstract: Have you ever encountered code with so many conditions and processing paths that it was almost impossible to maintain and extend? What if we replaced it with an automatically generated, self-improving algorithm? In recent years, machine learning as a field of artificial intelligence has become an effective tool for creating systems and applications. With the development of artificial neural networks, programming complex business rules and services based on prediction and classification can be replaced by pre-trained machine learning models. In this presentation, you will see a case study illustrating the potential of PHP in integrating machine learning. We will walk through the process of creating a classifier and placing it in a PHP-based project.

Lecture: Exploring the viability of PHP for implementing artificial neural networks: A case study on autonomous vehicle control with CNN model

Abstract: In recent years, machine learning has become an essential tool for developing intelligent systems. With the rise of artificial neural networks, programming languages such as Python and R have become the go-to options for machine learning implementation. But is PHP a viable alternative? In this presentation, we will explore the potential of PHP for implementing artificial neural networks by examining its limitations compared to other popular languages. We will also demonstrate the application of machine learning in PHP through a case study where we trained a convolutional neural network model to control a prototype of an autonomous vehicle using Raspberry Pi and Nvidia’s “DAVE 2” CNN model architecture.

Responsibilities:
– prepare laboratories for students related to computing vision recognition and training autonomus vehicle using convolutional neural network and tensorflow library
– assemble vehicles using Raspberry Pi 4 Model B, motors and other parts
– configure environment for model training and run model on Raspbian OS
– implement module for object detection

The “API in data mining” course aims to familiarize students with the tools, libraries and cloud solutions used in data mining. During the course, students will acquire the necessary skills to design complete applications that use cloud-based services to process and collect data. They will gain the ability to integrate and communicate between libraries and data mining tools. The contents of the module introduce you to the basic terms and concepts related to modern data processing pipeline architecture. After completing the course, participants will gain the basic skills necessary to use advanced tools and techniques in working with data and to design and implement cloud-based applications for data processing and collection.

topics:
Definition of API and data mining concepts, application of API in data mining
Cloud computing challenges,ETL and ELT
Specificity of the organization and roles related to working with data.
The structure of applications using machine learning models and big data-related services.
Integration with big data and machine learning services within AWS / Google Cloud AI.
Designing a data processing pipeline using websites to process and collect data.
Data import and preparation, data storage and structuring
Creating a data Lake, creating a data warehouse
Big Data processing
Scaling, containerization and microservices architecture in modern dev and prod env

The amount of data collected in online resources is growing at an exponential rate. To use the knowledge contained in this data, it is necessary to know machine learning techniques, which are part of the field of artificial intelligence. The module presents pattern mining techniques for both structured data, which follows a clearly defined schema, and unstructured data, which exists in the form of natural language text, signals, or graphics. Courses within the module include pattern detection, training machine learning models, clustering and grouping, text mining, computer vision processing and data analysis, and visualization. Practical exercises during laboratories include, among others: training a model for predicting apartment prices, training a prototype, an autonomous vehicle, using transfer learning to extract information from text, implementing sound and image recognition to control a drone.

topics:
• Python machine learning libraries • Supervised machine learning • Unsupervised machine learning • Reinforcement learning • Natural language processing • Detecting patterns in text data • Text tagging and classification • Topic modeling, semantic analysis • Artificial neural networks • Non-text data analysis: audio • Computer vision processing • Cloud frameworks and solutions in data science

Pictures from companyhouse.de

Search engine based on Elasticsearch

Responsibilities:

• setup multi-node Elasticsearch server structure
• implementing efficient synchronization script
• configuring queries and score functions