We use cookies

This website uses cookies to provide necessary website functionality, improve your experience and analyze our traffic. By using our website, you agree to our Privacy Policy and our cookies usage.

Our friendly team would love to hear from you.

Thank you for your interest!

We will contact you as soon as possible.

Back to homepage

"We would definitely recommend Cognitum as a IT partner in the field of AI, management systems, semantic technologies and software development."

Master NLP & LLMs with Cognitum course

Dive into the essentials of Natural Language Processing and Large Language Models. This course offers a hands-on approach to NLP workflows, tokenization, and the Hugging Face ecosystem, building your expertise step-by-step. Perfect for software engineers eager to master NLP.

Get started

Knowledge

Master NLP & LLMs in 5 days. Build a strong foundation in natural language processing — from tokenization to classification, embedding methods, and transformer architectures. Learn with hands-on exercises grounded in real-world use cases.

Skills

Practical experience with modern tools. Work directly with Huggingface, train and fine-tune models, and apply sequence-to-sequence architectures. Leave the course ready to build, adapt, and deploy real AI language solutions.

Network

Learn, share, and grow together. Join a vibrant group of developers, engineers, and AI enthusiasts. Exchange ideas, get feedback, and stay engaged through community-driven learning and collaborative challenges.

Session 1 – Introduction to NLP & LLMs

beginner

SESSION OVERVIEW

Day 1 – Tokenization & Introduction to NLP Models

Day 2 – Text Classification

Day 3 – Token Classification

Day 4 – Sequence-to-sequence models

Day 5 – Introduction to LLMs

Tokenization & Introduction to NLP Models

Agenda

Introduction to computer text representation
Tokenization
Language Modeling
Hugging Face Ecosystem

Target Participants

This day is ideal for beginners with no NLP experience, offering an introduction to tokenization, the Transformers library, and the Hugging Face ecosystem. Advanced users may also find the theoretical parts valuable for the following days of the course.

Text Classification

Agenda

Working closer with the datasets
Getting to know the most important metrics for binary classification
Learning about different ways of text classification

Target Participants

This part targets advanced participants working with neural network architecture in PyTorch, while also offering valuable insights for less experienced users through metrics, model fine-tuning, and dataset handling.

Token Classification

Agenda

Discussion of various real-world applications of Token Classification
Part of Speech Tagging
Named Entity Recognition (NER)
Optional task

Target Participants

This section is aimed at both beginner and advanced programmers, exploring new NLP tasks and use cases with the Transformers API. It also introduces monitoring tools and an evaluation library for those interested in training and model assessment.

Sequence-to-sequence models

Agenda

Examples of sequence-to-sequence problems
Full encoder-decoder architecture
Metrics used for sequence-to-sequence problems
Sequence-to-sequence models as universal NLP models
Text generation algorithms
Fine-tuning
Considerations regarding input/output lengths.

Target Participants

Building on the previous workshop, this session introduces new NLP tasks, evaluation metrics, and text generation methods relevant to Transformer architectures, offering value to participants at all levels.

Introduction to LLMs

Agenda

Introduction to using LLMs with transformers
Definition and showcase of zero-shot and few-shot learning techniques
Comparison of performance of the LLM and dedicated smaller models
Few-shot learning in detail
Text generation strategies in LLMs

Target Participants

This beginner-level workshop introduces LLMs, easing the transition from smaller models and gradually building a foundation for more advanced techniques.

Session 2 – Large Language Models

intermediate

SESSION OVERVIEW

Day 1 – Large Language Models Principles

Day 2 – Prompt Engineering Mastery

Day 3 – Agents, Benchmarks and Fine-tuning

Day 4 – Model Alignment

Day 5 – Efficient Inference

Large Language Models Principles

Agenda

Introduction to Large Language Models, revision and expansion
Basics of Prompt Engineering in detail
In-Context Learning

Target Participants

Designed for participants with limited LLM knowledge, this workshop is especially suitable for those who missed earlier sessions, while still offering valuable insights for more advanced attendees.

Prompt Engineering Mastery

Agenda

Improve Reasoning and Logic
Reduce Hallucination

Target Participants

While technically grounded in the Transformers library, this workshop emphasizes the theory behind prompt engineering, offering valuable insights for programmers of all experience levels.

Agents, Benchmarks and Fine-tuning

Agenda

LLMs as Agents
Benchmarking LLMs
Instruction fine-tuning

Target Participants

This part explores advanced technical topics and is recommended for participants with experience in ML frameworks like LangChain.

Model Alignment

Agenda

Efficient fine-tuning of LLMs
Model alignment

Target Participants

Maintaining the previous workshop’s difficulty, this session is designed for advanced participants focused on customizing LLM-based solutions.

Efficient Inference

Agenda

Quantization methods
Other approaches
vLLM

Target Participants

This workshop is recommended for those focused on building and deploying LLM-based systems, with prior experience in relevant tools being essential.

Session 3 – Retrieval Augmented Generation

advanced

SESSION OVERVIEW

Day 1 – RAG Demo

Day 2 – RAG Components & Evaluation

Day 3 – Improving Retrieval

Day 4 – Improving Generation

RAG Demo

Agenda

Rag overview
LangChain introduction
Retrieval fundamentals
Context-based generation
RAG Evaluation

Target Participants

This part introduces RAG, covering its components, advantages over standard LLM use, and ways to enhance its performance. The introductory day brings all participants up to speed, with the full session best approached as a comprehensive, step-by-step guide.

RAG Components & Evaluation

Agenda

Synthetic dataset generation
Vector database
Retrieval evaluation metrics

Target Participants

This part explores the retrieval component of RAG, covering alternative data sourcing, metadata enhancement, and separate evaluation methods.

Improving Retrieval

Agenda

Baseline implementation
Chunking
Hybrid Search
Cross-encoder
Bi-encoder

Target Participants

This part expands on earlier workshops by introducing new components and demonstrating how to tailor RAG systems—ideal for programmers seeking customization.

Improving Generation

Agenda

Evaluation metrics
Improving context
Fine-tuning the generator
Complex generation techniques
Security

Target Participants

This workshop, best suited for experienced RAG users, covers advanced components that enhance system usability, user-friendliness, and security.

Session 4 – Annotation, data and MLops

beginner

SESSION OVERVIEW

Day 1 – Data Annotation

Day 2 – Training Monitoring

Day 3 – Data-centric AI and Model Testing

Day 4 – Monitoring

Data Annotation

Agenda

Overview of open-source annotation tools
Data annotation
Evaluation

Target Participants

This day is dedicated to those aiming to streamline data collection—particularly annotation—though the broad scope ensures relevance for all.

Training Monitoring

Agenda

Introduction to MLflow
Training monitoring
Hyperparameter tuning with Optuna
Model registry

Target Participants

The workshop offers insights for all interested in neural networks, with key value for those directly involved in model training or fine-tuning.

Data-centric AI and Model Testing

Agenda

Text Data Quality Assurance
Model Quality Testing
Interpretability

Target Participants

This workshop focuses on optimizing data mining and post-training quality assurance. It benefits those committed to data quality and model interpretability, including data engineers and R&D researchers.

Monitoring

Agenda

Introduction to Monitoring
NLP Data Monitoring
NLP Model Monitoring and Concept Drift

Target Participants

This part addresses post-deployment actions, with key insights for data scientists and engineers involved in data preparation, model training, and fine-tuning—areas most impacted by data and concept drift.

Get Your Personalized
LLMs Course Quote

Interested in mastering Large Language Models? Fill out the form below to receive a tailored quotation for a course designed to meet your specific needs and objectives.

Thank you for your interest!

We will contact you as soon as possible.

Back to homepage

Sign Up For Our Newsletter

Creators/Trainers

Aleksander Smywiński-Pohl

NLP & LLM Expert

Krzysztof Wróbel

NLP & LLM Expert

Piotr Rybak

NLP & LLM Expert

Mykola Haltiuk

NLP & LLM Expert

Jakub Adamczyk

NLP & LLM Expert

FAQ

What are Large Language Models (LLMs)?

–

Large Language Models (LLMs) are a type of machine learning model for natural language processing. They’re trained on a large amount of text data and can generate human-like text based on the input they’re given.

What is a private LLM?

–

A private LLM is a large language model that is a model exclusively utilized by a specific organization. This guarantees data security and privacy as the model and its associated data are not disseminated to other entities.

Are LLMs secure?

–

Yes, especially when you use private LLMs. These models are not shared with other entities, ensuring your data remains secure and complies with your stringent data policies.

Can LLMs be integrated with existing systems?

–

Yes, LLMs can be seamlessly integrated with clients’ environments such as databases, websites, mobile apps, messaging apps, customer support platforms, and more.

How do I get started with implementing LLMs?

–

To start implementing LLMs, reach out to us at Cognitum. We’ll discuss your specific needs and how our solutions can help you achieve your goals.

What is a Generative AI Application?

–

A Generative AI application is a type of artificial intelligence that creates new content. It’s based on patterns and structure of their input training data and then generates new data.

Your certified partner!

Empower your projects with Cognitum, backed by the assurance of our ISO 27001 and ISO 9001 certifications, symbolizing elite data security and quality standards.

Master NLP & LLMs with Cognitum course

Knowledge

Skills

Network

Session 1 – Introduction to NLP & LLMs

Session 2 – Large Language Models

Session 3 – Retrieval Augmented Generation

Session 4 – Annotation, data and MLops

Get Your Personalized
LLMs Course Quote

Sign Up For Our Newsletter

Creators/Trainers

Aleksander Smywiński-Pohl

Krzysztof Wróbel

Piotr Rybak

Mykola Haltiuk

Jakub Adamczyk

FAQ

What are Large Language Models (LLMs)?

What is a private LLM?

Are LLMs secure?

Can LLMs be integrated with existing systems?

How do I get started with implementing LLMs?

What is a Generative AI Application?

Your certified partner!

Empower your projects with Cognitum, backed by the assurance of our ISO 27001 and ISO 9001 certifications, symbolizing elite data security and quality standards.

PN-EN ISO 9001:2015

ISO/IEC 27001

EN ISO 14001:2015

Master NLP & LLMs with Cognitum course

Knowledge

Skills

Network

Session 1 – Introduction to NLP & LLMs

Session 2 – Large Language Models

Session 3 – Retrieval Augmented Generation

Session 4 – Annotation, data and MLops

Get Your PersonalizedLLMs Course Quote

Sign Up For Our Newsletter

Creators/Trainers

Aleksander Smywiński-Pohl

Krzysztof Wróbel

Piotr Rybak

Mykola Haltiuk

Jakub Adamczyk

FAQ

What are Large Language Models (LLMs)?

What is a private LLM?

Are LLMs secure?

Can LLMs be integrated with existing systems?

How do I get started with implementing LLMs?

What is a Generative AI Application?

Your certified partner!

Empower your projects with Cognitum, backed by the assurance of our ISO 27001 and ISO 9001 certifications, symbolizing elite data security and quality standards.

PN-EN ISO 9001:2015

ISO/IEC 27001

EN ISO 14001:2015

Get Your Personalized
LLMs Course Quote