AI (Artificial Intelligence) integration training
AI (Artificial Intelligence) integration training
Durată: 3 zile
Certificare: Diploma de participare
Cui îi este dedicat cursul?
The training is addressed to the roles involved in the planning and adoption / integration of AI solutions:
● IT professionals
● Project Managers
● Other decision-makers
The potential participants will want to:
● Understand the AI integration possibilities - opportunities, challenges, benefits
● Understand the integration roadmap
● Define a series of integration milestones
Cunoștințe și abilități inițiale
The following is a list of the desireable prerequisites required to attend the course:
For technical audience (IT professionals):
● Mandatory:
1. Technical Background:
○ Solid understanding of software development processes
○ At least one programming language (preferably Python, given its prominence in AI)
2. Basic knowledge of AI and Machine Learning:
○ Familiarity with the concepts covered in the "AI Introduction" course or equivalent knowledge
○ Understanding AI use-cases and fundamental principles
3. Data management understanding: data handling knowledge, including:
○ Databases, data preprocessing
○ The importance of data quality
● Optional:
○ Familiarity with cloud computing:
■ An understanding of cloud services is beneficial, but not required
■ Familiarity with deploying & managing applications on platforms like AWS, Google Cloud, or
Azure will be helpful
○ Business acumen: while not mandatory for technical participants, an understanding of the business
context where AI is applied will enhance the ability to integrate solutions effectively
For Project Management audience:
● Mandatory:
1. Basic technical literacy:
a. Familiarity with the basics of software development
b. The ability to understand technical language and concepts
2. Project management skills: strong Project Management background, with an emphasis on technology
or IT projects
3. Basic knowledge of AI and Machine Learning: a high-level understanding of what AI and ML are,
including potential use cases and implications for businesses
● Optional:
○ Cloud computing familiarity: understanding cloud computing basics is helpful for managing projects
involving cloud-based AI solutions, but is not required for foundational knowledge
○ Technical background: while not necessary, some programming concepts knowledge or data
management will greatly help in:
■ The communication with technical teams
■ The understanding of project scopes more deeply
Prezentarea cursului
The training duration is 3 days, 6 or 7 hours each day. The scheduling per days:
● Day 1: AI foundations and enterprise strategy
● Day 2: Technical deep dive and implementation
● Day 3: Scaling and evolving AI, AI integration workshop
Sessions scheduling:
● Each training day is composed from 6 or 7 hours, depending on the agreed duration
● We will have a break at each ~50 minutes. Depending on their complexity and on the questions and
discussions, some sessions may take more or less than 50 minutes
Ce subiecte abordează cursul
Day 1: AI foundations and enterprise strategy
1. Training overview
2. Introduction to enterprise AI
○ Basics of AI and ML (needed only if the participants haven’t attended the ‘AI Introduction’ course,
shortened otherwise)
■ AI intro and reasoning - high level overview
■ Types of AI and ML - short overview / recap
■ AI models overview
● Foundation models
● Pre-trained models repos (ex: HuggingFace)
○ AI use-cases in enterprises
■ Overview of the most common use-cases
■ AI integration templates
○ Assessing AI readiness
■ Technical
■ Financial
■ Cultural & social acceptance
■ GDPR, data integrity, other legal aspects
3. AI project lifecycle in enterprises
○ Understanding the AI lifecycle - working with AI models
■ What is a pretrained model?
■ Why to use pretrained models?
■ Identify the most useful models
■ How are pretained models advancing AI?
■ Model repos overview
■ Train with business specific (/ internal) datasets
■ Evaluate
■ How to fine tune a model?
○ Introduction to MLOps
○ Overview of the main technologies
■ TensorFlow
■ PyTorch
■ Keras
4. Workshop on AI strategy
○ Identifying opportunities
■ Analyze business needs
■ Find relevant data sources
○ Hands-on work with pre-trained models - alternatives (established with the beneficiary company):
■ Using HuggingFace models / LLMs / transformers etc
● Google Colab
● HuggingFace spaces
■ Local downloaded LLM
■ Local Docker container or VM, downloadable on each user's machine
■ Remote VM
● Per user
● For everyone
5. Group discussion
○ Sharing of experiences and expectations
○ Q&A session
Day 2: Technical deep dive and implementation
1. Data management for AI integrations
○ Data strategies and governance
○ Vector embeddings and DBs
■ Vector embeddings definition
■ Types of vector embeddings
■ Applications of vector embeddings
■ Vector databases providers
■ The implementation of an example solution
2. Building AI solutions
○ Algorithms and practical applications
○ Retrieval-Augmented Generation (RAG)
■ What is Retriever-Augmented Generation (RAG)
■ RAG diagram
■ The retriever: finding relevant knowledge
■ The generator: crafting the response
■ Business use cases
○ LangChain - bridging language models with external knowledge
■ Overview
■ Integration
3. AI integration and deployment
○ Model integration in existing software systems
○ Virtualization and containerization
■ Docker overview
■ Kubernetes and OpenShift overview
○ CI&CD pipelines
■ Overview
■ CI&CD solutions overview - Jenkins, ArgoCD, Spinnaker etc
4. Workshop on cloud and AI
○ Cloud platforms overview: AWS, Azure, Google Cloud
○ Serverless architectures
○ Cloud AI services
○ Hands-on with Jupyter Notebooks / Google Colab
Day 3: Scaling and evolving AI, AI integration workshop
1. Scaling AI in the enterprise
○ Scaling strategies: from PoC to full-scale deployments
○ Performance tuning and optimization
○ Monitoring and maintenance of AI systems
2. Emerging trends and technologies
○ Advanced applications of pre-trained models
○ Future of RAG and Langchain
○ Quantum computing in AI
3. AI integration workshop
○ Hands-on project work, on two teams - to be defined with the participants
○ Solutions presentations - each team will present their work, challenges and learnings
○ Q&A session
4. Course feedback
