Prerada prirodnog jezika (NLP) Training in Belgrade

This live, instructor-guided training in Belgrade empowers finance practitioners and IT staff to implement Ollama locally. Learners acquire the skills to deploy, configure, and embed large language models into secure financial workflows, while adhering to best practices for compliance and automation.

This live training in Belgrade empowers healthcare practitioners and IT teams to deploy Ollama for local LLMs. Participants will learn to integrate AI into clinical workflows, customize models for specific tasks, and maintain strict data privacy and regulatory compliance within secure healthcare environments.

This live, instructor-led session on Belgrade focuses on applying responsible AI principles within Ollama, with an emphasis on fairness, transparency, and accountability. Learners will create governance workflows, deploy bias mitigation strategies, and establish compliance monitoring frameworks through practical labs and real-world case studies.

This hands-on training in Belgrade is targeted at intermediate to advanced professionals seeking to build, deploy, and manage AI agents. Participants will learn to utilize Mistral Medium 3, Le Chat Enterprise, and Devstral for robust enterprise and developer use cases, including the integration of connectors, RAG patterns, and compliance measures.

This live, instructor-led training in Belgrade empowers intermediate professionals with the expertise to deploy Ollama securely. Participants will learn to apply differential privacy, implement secure logging, and enforce data access controls that align with regulatory compliance requirements, ensuring robust on-premises environments.

Ova uživo obuka na Belgrade pokriva integraciju Mistral konverzacijskih asistenata u preduzećke radne tokove. Učesnici će naučiti kako da povežu SaaS alate, implemetiraju generisanje uz pomoć pretrage i dizajniraju efikasne UX obrasce za chat interfejse usmerene na korisnike i interne timove.

This instructor-led, live training in Belgrade helps beginners apply Claude AI to customer support and service operations. Participants will learn to automate triage, draft tone-consistent responses, and integrate AI into existing workflows using practical, hands-on exercises.

Gain mastery over LangGraph for legal applications through this comprehensive 35-hour instructor-led course in Belgrade. Designed for intermediate to advanced professionals, the curriculum covers workflow design, compliance controls, and human-in-the-loop approval processes. You will learn to build auditable, production-ready LLM solutions with confidence.

Designed specifically for compliance and security professionals, this live Mistral Responsible AI training in Belgrade equips you with the skills to manage privacy-preserving techniques, data residency strategies, and enterprise controls like RBAC and SSO, ensuring a secure and compliant AI deployment.

This live Claude AI training in Belgrade empowers professionals to integrate the tool into data analysis and BI tasks. It addresses exploratory data analysis, SQL generation, and dashboard design through natural language prompting, with the goal of streamlining workflows and enhancing the extraction of insights from business data.

This instructor-led training in Belgrade empowers professionals to create LangGraph-based finance solutions. Learners will master implementing workflows that align with regulatory requirements, integrating data standards, and deploying systems with robust governance, observability, and compliance controls.

Gain the skills to construct multimodal applications leveraging Mistral models, with a focus on OCR, document intelligence, and vision-text search. This practical training covers pipeline execution, NLP integration, and enterprise-grade architecture, delivered either online or in Belgrade.

This instructor-led, live training in Belgrade (online or onsite) targets intermediate-level data scientists and developers who aim to apply NLP techniques using Python in Google Colab.

By the end of this training, participants will be able to:

• Understand the core concepts of natural language processing.
• Preprocess and clean text data for NLP tasks.
• Perform sentiment analysis using NLTK and SpaCy libraries.
• Work with text data using Google Colab for scalable and collaborative development.

This advanced, instructor-led training on Belgrade centers on diagnosing, testing, and evaluating local Ollama setups. The curriculum addresses systematic debugging, robust evaluation pipelines, observability, and automated CI/CD integration to secure production reliability and operational readiness.

This instructor-led live training in Belgrade (online or onsite) targets intermediate-level researchers, analysts, and knowledge workers who wish to leverage Claude AI for processing large data volumes, synthesizing information, and automating research tasks.

By the end of this training, participants will be able to:

• Use Claude AI for literature reviews and research synthesis.
• Summarize reports, research papers, and long-form documents.
• Extract key insights and trends from structured and unstructured data.
• Integrate Claude AI into research and knowledge management workflows.

Ova obuka uživo, pod vodstvom instruktora u oblasti Belgrade, namenjena je inženjerima mašinskog učenja koji žele da usvoje samostalno hostovanje i prilagođavanje modela Devstral i Mistral. Učesnici uče praktične tehnike upravljanja, monitoringa i usklađenosti kako bi samouvereno implementirali veštačku inteligenciju otvorenog koda u proizvodnim okruženjima.

Discover the application of LangGraph in healthcare through this live training held in Belgrade, designed for intermediate to advanced professionals. Learn to engineer stateful, compliant workflows, integrate standards like FHIR and SNOMED CT, and navigate ethical and operational challenges within regulated environments.

This instructor-led, live training in Belgrade is tailored for advanced AI platform engineers, AI DevOps specialists, and ML architects aiming to optimize, debug, monitor, and operate production-grade LangGraph systems, which build stateful, multi-actor LLM applications as composable graphs.

Ova uživa obuka sa predavačem, održavana u <lok> (online ili uživo), namenjena je kreatorima sadržaja, autorima i marketinškim stručnjacima početničkog do srednjeg nivoa znanja, koji žele da iskoriste Claude AI za generisanje, uređivanje i sažimanje sadržaja kako bi poboljšali blogove, izveštaje i marketinške materijale.

Na kraju ove obuke, učesnici će moći da:

• Koriste Claude AI za buru ideja i proširenje koncepata.
• Poboljšaju pisanje uz pomoć AI-asistiranih uredničkih i pravopisnih korekcija.
• Generišu sažetke za duge tekste i izveštaje.
• Automatizuju kreiranje marketinških tekstova za različite platforme.

This instructor-led, live training in Belgrade (online or onsite) is aimed at advanced-level professionals who wish to fine-tune and customize AI models on Ollama for enhanced performance and domain-specific applications.

By the end of this training, participants will be able to:

• Set up an efficient environment for fine-tuning AI models on Ollama.
• Prepare datasets for supervised fine-tuning and reinforcement learning.
• Optimize AI models for performance, accuracy, and efficiency.
• Deploy customized models in production environments.
• Evaluate model improvements and ensure robustness.

Develop expertise in cost-efficient Mistral LLM architectures through this 14-hour live training session in Belgrade. Acquire the skills to deploy, optimize, and scale Mistral Medium 3, maximizing throughput while minimizing costs by leveraging batching, quantization, and high-efficiency serving strategies for enterprise needs.

This instructor-led, live training in Belgrade (online or onsite) is aimed at advanced-level professionals who wish to implement secure and efficient AI-driven workflows using Ollama.

By the end of this training, participants will be able to:

• Deploy and configure Ollama for private AI processing.
• Integrate AI models into secure enterprise workflows.
• Optimize AI performance while maintaining data privacy.
• Automate business processes with on-premise AI capabilities.
• Ensure compliance with enterprise security and governance policies.

This instructor-led, live training in Belgrade (online or onsite) is aimed at intermediate-level software developers and AI engineers who wish to integrate Claude AI into their applications, build AI-powered chatbots, and enhance software functionality with AI-driven automation.

By the end of this training, participants will be able to:

• Use the Claude AI API to integrate AI into applications.
• Develop AI-driven chatbots and virtual assistants.
• Enhance applications with AI-powered automation and NLP.
• Optimize and fine-tune Claude AI models for different use cases.

This live training in Belgrade introduces Le Chat Enterprise, a private ChatOps platform for secure, governed conversational AI. Learners will master the deployment, configuration, and integration of this tool with enterprise apps, activating RBAC, SSO, and compliance-driven controls to build scalable and secure AI workflows.

Become proficient in Devstral through this live training session held in Belgrade. The curriculum covers architecture, agentic workflow design, and tool integration. Designed for advanced engineers, the course utilizes hands-on labs to construct coding agents that elevate both productivity and security within contemporary development environments.

This instructor-led, live training in Belgrade (online or onsite) is aimed at beginner-level professionals who wish to integrate Claude AI into their daily workflows to improve efficiency and automation.

By the end of this training, participants will be able to:

• Use Claude AI for automating repetitive tasks and streamlining workflows.
• Enhance personal and team productivity using AI-powered automation.
• Integrate Claude AI with existing business tools and platforms.
• Optimize AI-driven decision-making and task management.

This instructor-led, live training in Belgrade (online or onsite) is aimed at intermediate-level professionals who wish to deploy, optimize, and integrate LLMs using Ollama.

By the end of this training, participants will be able to:

• Set up and deploy LLMs using Ollama.
• Optimize AI models for performance and efficiency.
• Leverage GPU acceleration for improved inference speeds.
• Integrate Ollama into workflows and applications.
• Monitor and maintain AI model performance over time.

This live course, available online or onsite Belgrade, guides intermediate engineers in constructing dynamic LLM workflows with LangGraph. Participants will learn to design stateful graphs, apply branching logic, and integrate retrieval, APIs, and memory to build robust, context-aware applications.

This instructor-led, live training in Belgrade (online or onsite) is aimed at intermediate-level data scientists, AI developers, and AI enthusiasts who wish to use LLMs to perform various NLP tasks and create novel and diverse content for different purposes.

By the end of this training, participants will be able to:

• Establish a development environment with LLMs and essential tools.
• Expertly perform NLU and NLI tasks with LLMs.
• Extract, infer, and utilize knowledge graphs effectively.
• Generate and manage dialogues using LLMs for conversational applications.
• Evaluate content quality and diversity generated by LLMs and generative AI.
• Apply ethical principles, ensuring fairness and responsible use of LLMs.

This instructor-led, live training in Belgrade (online or onsite) is aimed at intermediate-level professionals who wish to enhance their NLP projects through the effective fine-tuning of pre-trained language models.

By the end of this training, participants will be able to:

• Grasp the core concepts behind fine-tuning for NLP applications.
• Apply fine-tuning techniques to pre-trained models like GPT, BERT, and T5 for targeted NLP use cases.
• Adjust hyperparameters to boost model effectiveness.
• Assess and implement fine-tuned models in practical, real-world settings.

This instructor-led LangGraph Fundamentals course in Belgrade empowers developers to construct graph-based LLM applications. You will acquire proficiency in core concepts such as nodes, edges, state management, and branching logic. The curriculum guides you through integrating tools, retrieval systems, and external APIs, culminating in hands-on labs for testing and deploying reliable workflows.

This instructor-led training in Belgrade helps advanced AI engineers deploy and secure Mistral Medium 3. You will optimize inference costs, implement multimodal use cases, and apply enterprise compliance practices. Includes hands-on labs for API and self-hosted setups.

This instructor-led, live training in Belgrade (online or onsite) targets beginner-level professionals aiming to install, configure, and use Ollama for running AI models on their local machines.

Upon completion of this training, participants will be able to:

• Grasp the core principles and capabilities of Ollama.
• Configure Ollama for executing local AI models.
• Deploy and interact with LLMs using Ollama.
• Enhance performance and manage resources for AI workloads.
• Examine application scenarios for local AI deployment across various sectors.

This two-day course is designed for developers who already utilize AI coding tools like Cursor, Claude Code, or Copilot. It focuses on delegating complete tasks to agents, constructing the personalization stack (Rules, AGENTS.md, Skills, MCP, Agents), connecting and developing MCP servers, executing agents in parallel, and implementing a structured agentic workflow. The course material is tool-agnostic and builds upon the Foundations course.

A two-day introductory course designed for individuals new to AI-assisted software development, including those with minimal or no coding background. The program explores the underlying mechanics of these tools, strategies for directing them via prompts, and practical applications for routine tasks such as building small projects from the ground up, navigating existing codebases, and evaluating outcomes. The content is tool-agnostic, relevant to platforms like Cursor, Claude Code, Copilot, and comparable solutions. This course serves as a precursor to the 'Agentic AI Development: Advanced Workflows' program.

This two-stage training program is tailored for software delivery teams aiming to embed AI-assisted engineering practices into their routine operations by leveraging Claude Code.

This course at Belgrade provides Artificial Intelligence training covering data analysis, distributed machine learning, and natural language processing, designed to help developers build large-scale intelligent systems. The content explores practical methods for Spark MLlib, recommendation engines, Word2Vec semantic analysis, and RNN-LSTM architectures, enabling professionals to convert raw data into automated decision-making pipelines deployable in production environments, thereby comprehensively enhancing core competencies in building intelligent systems

This live, instructor-led training in Belgrade is tailored for intermediate-level developers seeking to design, build, and test AI agents utilizing the Claude API and the Agent SDK.

Upon completion, participants will be equipped to:

• Invoke the Messages API utilizing system prompts, structured outputs, and streaming capabilities
• Configure tools and execute a robust tool-use cycle
• Integrate agents with external systems via MCP servers
• Construct and operate agents using the Claude Agent SDK, incorporating permission management and subagents
• Optimize costs and latency through prompt caching and effective context management
• Evaluate, record, and monitor agent performance using a concise evaluation dataset

This live, instructor-led training in Belgrade is designed for intermediate to advanced developers and tech leads seeking to configure and scale Claude Code for consistent team-level utilization.

By the conclusion of the course, participants will have the capability to:

• Structure CLAUDE.md files, settings, and permissions for specific repositories and teams
• Create specialized subagents with focused roles and defined tool access
• Implement hooks that automatically enforce formatting, testing, and safety rules
• Package recurring workflows into Skills, slash commands, and plugins
• Link Claude Code to external systems via MCP servers
• Run Claude Code non-interactively in scripts and CI pipelines, and critically assess its output

This live, instructor-led session in Belgrade is tailored for business users who wish to construct simple internal tools and prototypes.

By the end of the training, participants will have the ability to:

• Describe a small tool sufficiently for AI to build it
• Develop a working prototype using Lovable
• Utilize Claude Code for creating simple helper tools with guidance
• Determine when a prototype necessitates developer input

This instructor-led live training in Belgrade caters to knowledge workers and managers aiming to delegate entire projects to Claude.

By the end of the session, attendees will be capable of:

• Identifying the independent functions of Claude Cowork.
• Outsourcing research, document handling, and file operations.
• Planning, auditing, and correcting delegated outcomes.
• Recognizing limitations and ensuring secure usage.

Claude Cowork preuzima celokupne zadatke, poput istraživanja, pripreme dokumenata i rada sa fajlovima, umesto da odgovara na jedan prompt odjednom. Ovaj kurs uči kako da se sastave kvalitetne zadatke, kako da se proverava plan i kako da se pregledaju rezultati. Učešnici preuzmu tri stvarna zadatka i podesavaju zadatka sa rasporedom izvođenja.

Ova obuka sa predavačem uživo u Belgrade namenjena je: radnicima znanja, asistentima, vodjama projekata i odeljenja koji žele da poveruju ceo zadatak Claude-u, umesto da komuniciraju korak po korak.

Na kraju ove obuke, učesnici će biti u mogućnosti da:

• razumeju razliku između dijaloga i delegiranog rada u Cowork-u
• sastave zadatke sa jasnim ciljem, materijalima, ograničenjima i kriterijumima za prihvatanje
• izrađuju dokumente, elektronske tabele i prezentacije iz određene mape
• delegiraju istraživanje od postavljanja pitanja do završenog izveštaja
• koriste pluginove, veštine i konektore u Cowork-u
• definišu tačke odobravanja, rezervne kopije i bezbedan pristup mapama

Format kursa

• Trajanje: 1 dan (7 sati)
• Interaktivna obuka sa trenerom, diskusija i praktične vežbe na osnovu sopstvenih slučajeva upotrebe učesnika
• Rezultat, 1. dan: Svaki učesnik je poverio tri stvarna zadatka Cowork-u, pregledao rezultate i podesio zadatak sa rasporedom izvođenja.

Opcije prilagođavanja kursa

• Da biste zatražili prilagođenu obuku za ovaj kurs, molimo vas da se javite nama radi koordinacije.

Ollama je alat otvorenog koda za pokretanje modela velikog jezika lokalno na potrošačkoj i preduzećkoj opremi. On apstrahuje kvantizaciju modela, alokaciju GPU procesora i posluživanje API-ja u jedinstveno komandno sučelje, omogućavajući organizacijama da samostalno hostuju modele poput Llama, Mistral i Qwen, bez slanja upita ili podataka OpenAI, Anthropicu ili Googleu.