NEW SPECIALIZATION
Building Next-Gen AI Solutions
with Agents
Learn to design and deploy intelligent multi-agent systems that leverage LLMs to solve real-world challenges at scale.
Enroll now
ELEVATE YOUR EXPERTISE
ELEVATE YOUR EXPERTISE
ELEVATE YOUR EXPERTISE
About the specialization
About the specialization
About the specialization
This specialization will guide you through the practical development of LLM-powered agents, from basic prompt engineering to production-level multi-agent systems. You will build hands-on projects, explore core architectures (RAG, memory/state management, multi-agent collaboration), and learn how to deploy robust agent solutions at scale.
By the end of this program, you will:
• Understand how to incorporate external data sources into LLMs.
• Build agents capable of complex reasoning and tool usage.
• Manage memory, state, and multi-agent workflows effectively.
• Deploy and monitor agent systems in production environments.
This specialization will guide you through the practical development of LLM-powered agents, from basic prompt engineering to production-level multi-agent systems. You will build hands-on projects, explore core architectures (RAG, memory/state management, multi-agent collaboration), and learn how to deploy robust agent solutions at scale.
By the end of this program, you will:
• Understand how to incorporate external data sources into LLMs.
• Build agents capable of complex reasoning and tool usage.
• Manage memory, state, and multi-agent workflows effectively.
• Deploy and monitor agent systems in production environments.
This specialization will guide you through the practical development of LLM-powered agents, from basic prompt engineering to production-level multi-agent systems. You will build hands-on projects, explore core architectures (RAG, memory/state management, multi-agent collaboration), and learn how to deploy robust agent solutions at scale.
By the end of this program, you will:
• Understand how to incorporate external data sources into LLMs.
• Build agents capable of complex reasoning and tool usage.
• Manage memory, state, and multi-agent workflows effectively.
• Deploy and monitor agent systems in production environments.
NEW SPECIALIZATION
Building Next-Gen AI
Solutions with Agents
June 3rd, 2025
Learn to design and deploy intelligent multi-agent systems that leverage LLMs to solve real-world challenges at scale.
Enroll now
NEW SPECIALIZATION
Building Next-Gen AI
Solutions with Agents
June 3rd, 2025
Learn to design and deploy intelligent multi-agent systems that leverage LLMs to solve real-world challenges at scale.
Enroll now
NEW SPECIALIZATION
Building Next-Gen
AI Solutions with
Agents
June 3rd, 2025
Learn to design and deploy intelligent multi-agent systems that leverage LLMs to solve real-world challenges at scale.
Enroll now
NEW SPECIALIZATION
Building Next-Gen
AI Solutions with
Agents
June 3rd, 2025
Learn to design and deploy intelligent multi-agent systems that leverage LLMs to solve real-world challenges at scale.
Enroll now
HIGH-LEVEL CONTENT OVERVIEW
HIGH-LEVEL CONTENT OVERVIEW
Program content
Program content
Maser the Coding interview.
01
01
From Prompts to RAG
From Prompts to RAG
Prompt-only models limitations
What is RAG?
Vector databases
Retrieval optimization for accurate result
Building a basic support agent
Prompt-only models limitations
What is RAG?
Vector databases
Retrieval optimization for accurate result
Building a basic support agent
02
From RAG to Agents
Why Agents?
Taxonomy of an Agent
Profiling
Memory
Planning
Actions
02
From RAG to Agents
Why Agents?
Taxonomy of an Agent
Profiling
Memory
Planning
Actions
Key Topics & Concepts
Dynamic Data: Prices, product reviews, or any data that frequently changes.
Why Agents?
• Multi-step reasoning about tradeoffs.
• Ability to integrate new information on the fly.
Agent Architecture:
• Orchestrating retrieval, transformation, and final answers.
03
03
Tools & LangGraph
Tools & LangGraph
Introduction to LangGraph, Graph Structure: nodes, edges and state management
Tool Usage: Searching external APIs, analyzing data, formatting outputs.
LLM integrations: Model Context Protocol (MCP)
Introduction to LangGraph, Graph Structure: nodes, edges and state management
Tool Usage: Searching external APIs, analyzing data, formatting outputs.
LLM integrations: Model Context Protocol (MCP)
04
Memory & State Management
Memory Architectures
Working Memory
Long-Term Memory
Implementing a Meeting Assistant Agent
04
Memory & State Management
Memory Architectures
Working Memory
Long-Term Memory
Implementing a Meeting Assistant Agent
Key Topics & Concepts
Graph Structure
• Nodes: Distinct operations (e.g., web search, analysis).
• Edges: Flow between operations, including conditional
paths.
• State Management: TypedDict usage across nodes.
Tool Usage:
• Searching external APIs, analyzing data, formatting outputs.
LangGraph:
• Building explicit workflows with maintainable state transitions.
05
05
Multi-Agent Systems
Multi-Agent Systems
Multi-Agent Architecture
Networked Agents
Supervisor/Worker Model
Collaboration & State Sharing
Multi-Agent Architecture
Networked Agents
Supervisor/Worker Model
Collaboration & State Sharing
06
Advanced Agent Patterns
Router Pattern
Parallelization Strategies
Human-in-the-Loop
06
Advanced Agent Patterns
Router Patter
Parallelization Strategies
Human-in-the-Loop
Key Topics & Concepts
Router Pattern:
• Intelligent classification of tasks (e.g., ticket routing).
• Using LLMs to decide how tasks are distributed.
Parallelization Strategies
• Map-reduce approach for handling large batches.
• Managing concurrency and collating results.
Human-in-the-Loop
• When to pause for manual decisions or validation.
• Resuming flows with updated state.
07
07
Agent Frameworks
Agent Frameworks
LangGraph
AutoGen
Smolagents
Framework Selection: Trade-offs and considerations
LangGraph
AutoGen
Smolagents
Framework Selection: Trade-offs and considerations
08
Production Deployment
Production Architecture
Monitoring & Cost Control
Security Considerations
08
Production Deployment
Production Architecture
Monitoring & Cost Control
Security Considerations
Key Topics & Concepts
Production Architecture
• Queue-based processing, load balancing.
• Ensuring consistent state across instances.
Monitoring & Cost Control
• Tracking usage costs per request.
• Performance metrics and error handling.
Security Considerations
• Rate limiting, input validation, and access control.
• Handling sensitive data with caution.
Final Project
Final Project
AI Shopping Assistant
AI Shopping Assistant
Design and build a conversational agent capable of handling both sales and customer support flows, including semantic and structured product search, cart management, agent routing, and human escalation, all integrated via tool use and memory.
Design and build a conversational agent capable of handling both sales and customer support flows, including semantic and structured product search, cart management, agent routing, and human escalation, all integrated via tool use and memory.
Enroll now
Enroll now
HIGH-LEVEL CONTENT OVERVIEW
HIGH-LEVEL CONTENT OVERVIEW
Program content
01
From Prompts to RAG
Prompt-only models limitations
What is RAG?
Vector databases
Retrieval optimization for accurate result
Building a basic support agent
02
From RAG to Agents
Why Agents?
Taxonomy of an Agent
Profiling
Memory
Planning
Actions
03
Tools & LangGraph
Introduction to LangGraph, Graph Structure: nodes, edges and state management
Tool Usage: Searching external APIs, analyzing data, formatting outputs.
LLM integrations: Model Context Protocol (MCP)
04
Memory & State Management
Memory Architectures
Working Memory
Long-Term Memory
Implementing a Meeting Assistant Agent
05
Multi-Agent Systems
Multi-Agent Architecture
Networked Agents
Supervisor/Worker Model
Collaboration & State Sharing
06
Advanced Agent Patterns
Router Patter
Parallelization Strategies
Human-in-the-Loop
07
Agent Frameworks
LangGraph
AutoGen
Smolagents
Framework Selection: Trade-offs and considerations
08
Production Deployment
Production Architecture
Monitoring & Cost Control
Security Considerations
Final Project
AI Shopping Assistant
Design and build a conversational agent capable of handling both sales and customer support flows, including semantic and structured product search, cart management, agent routing, and human escalation, all integrated via tool use and memory.
Enroll now
TIME COMMITMENT
Specialization schedule
Specialization takes place on Tuesdays and Thursdays.
Enroll now
TIME COMMITMENT
Specialization schedule
Specialization takes place on Tuesdays and Thursdays.
Enroll now
FIND OUT
FIND OUT
FIND OUT
How it works
How it works
How it works
Live sessions
with experts
Live sessions with experts
Live sessions with experts
You will be able to get guidance from experienced instructors during live interactive sessions.
You will be able to get guidance from experienced instructors during live interactive sessions.
Flexible
Schedule
Flexible Schedule
Flexible Schedule
The program schedule is 4-week duration with 8 live sessions (32 hours total). Two 2-hour sessions per week.
The program schedule is 4-week duration with 8 live sessions (32 hours total). Two 2-hour sessions per week.
Eligibility
criteria
Eligibility criteria
Eligibility criteria
You must be a Graduated Fellow at Anyone AI and fully up to date with all contractual commitments.
You must be a Graduated Fellow at Anyone AI and fully up to date with all contractual commitments.
Get hands-on experience
Get hands-on experience
Get hands-on experience
You will build skills through hands-on learning with world-class instructors' support.
You will build skills through hands-on learning with world-class instructors' support.
TIME COMMITMENT
TIME COMMITMENT
Specialization schedule
Specialization schedule
Specialization takes place on Tuesdays and Thursdays.
Enroll now
Enroll now
LEARN WITH EXPERTS
LEARN WITH EXPERTS
LEARN WITH EXPERTS
About our instructor
About our instructor
About our instructor
Background
Background
Background
Msc. in Data Science from Universidad de Buenos Aires. AI Engineer with 5+ years of experience in Data Science, Machine Learning, LLMs, NLP, and multi-agent architectures, specializing in autonomous AI systems, RAG pipelines.
Expertise
Expertise
Expertise
He focuses on LLMs, NLP, and autonomous agent systems. He has led end-to-end AI implementations, designing multi-agent architectures, scalable LLM-powered applications, and RAG pipelines for document processing and automation. Beyond industry work, he has contributed to academic research and teaching Natural Language Processing (NLP) at leading universities.
NEVER STOP LEARNING
NEVER STOP LEARNING
The time is now — enroll today!
The time is now — enroll today!
NEVER STOP LEARNING
The time is now — enroll today!
NEVER STOP LEARNING