Agentic AI
Engineer

• Introduction to text embeddings and semantic similarity
• Embedding models: OpenAI, Cohere, HuggingFace
• Understanding vector dimensions and inference performance
• Comparing cosine similarity vs Euclidean distance
• Embedding model selection and best practices

Vector Databases

• Introduction to FAISS, Pinecone, Chroma, Weaviate
• Indexing strategies
• Vector CRUD operations and metadata filtering
• Performing semantic search over chunked documents

RAG Implementation

• Chunking, indexing, and context injection
• LangChain retriever + prompt + LLM orchestration

No-Code Platforms

• Overview of no-code platforms for GenAI development
• Rapid prototyping with visual builders
• Low-code/no-code RAG implementations
• Comparing different no-code platforms

RAG with LangFlow

• LangFlow overview and visual no-code workflow design
• Node types: retriever, prompt, LLM, output
• Building RAG pipelines visually
• Combining structured and unstructured tools
• Exporting and deploying LangFlow projects

• Introduction to AWS Bedrock and foundation models
• Using Titan, Claude, and other models via Bedrock
• Setting up and securing a RAG workflow in Bedrock
• Integration with AWS services: S3, RDS, DynamoDB
• Monitoring cost and performance optimization
• Best practices for production RAG on AWS

• What is Agentic AI? Understanding Tools, Memory, and Planning
• Agent architecture and decision-making workflows
• Agent types: ReAct, Function Calling, Tool-using agents
• Agent reasoning loops and execution patterns
• When to use agents vs traditional RAG
• Agent use cases across industries

• Introduction to Agno Framework
• Building functional agents with Agno
• Tool integration and execution patterns
• Implementing agent memory and state management
• Agent debugging and testing strategies
• Advanced agent patterns with Agno

• Introduction to CrewAI framework
• CrewAI architecture: roles, tasks, and collaborative agents
• Designing multi-agent systems and workflows
• Assigning memory and tools to individual agents
• Simulating agent-based research and review processes
• Agent coordination and communication patterns
• Evaluating multi-agent output quality

Agentic AI with LangFlow

• Using LangFlow to wire agents visually
• Creating custom tools and chaining agent steps
• Building multi-turn conversational agents with memory
• Agent workflow design and optimization
• Exporting and reusing LangFlow agent configurations

No-Code Platforms for Agents

• No-code agent builders and platforms
• Comparing LangFlow with other no-code solutions
• Best practices for visual agent development

FastAPI for Production

• Building production-ready APIs with FastAPI
• Request/response handling and validation
• Authentication and authorization
• API documentation with Swagger/OpenAPI

Docker & Cloud Run

• Containerization fundamentals with Docker
• Creating Dockerfiles for GenAI applications
• Deploying containerized applications to Cloud Run
• Serverless scaling and CI/CD pipelines
• Cost optimization strategies on Cloud Run

LangSmith for Monitoring

• Logging prompt input/output, latency, and user feedback
• Tracing tool calls and memory steps in LangChain apps
• Evaluating hallucinations and response grounding
• Debugging and troubleshooting with LangSmith

LangWatch for Improvement

• Live monitoring and real-time analytics
• Prompt scoring and quality metrics
• User feedback collection and analysis
• Custom evaluation functions (RAGAS, MT-Bench)

Model Context Protocol (MCP)

• Understanding MCP architecture and design principles
• Connecting AI models to external data sources
• Building MCP servers and clients
• Use cases: database integration, API connections, tool access

Agent-to-Agent (A2A) Protocols

• Introduction to A2A communication standards
• Inter-agent communication patterns and protocols
• Message passing and coordination mechanisms
• Integration with CrewAI, LangChain frameworks
• Handling agent failures and error recovery

AI Safety Mechanisms

• Content filtering and moderation strategies
• Prompt injection prevention techniques
• Output validation and sanitization
• Implementing Guardrails AI & NeMo Guardrails

LLM Guard & Security

• Introduction to LLM Guard
• Input/output validation and security testing
• Adversarial attack prevention and mitigation
• Jailbreak prevention strategies

ML Fundamentals

• Overview of traditional ML models vs GenAI
• Supervised, unsupervised, and reinforcement learning
• Common ML algorithms: regression, classification, clustering
• When to use ML vs LLMs · Hybrid approaches

ML Use Cases

• Predictive analytics and forecasting
• Anomaly detection and pattern recognition
• Recommendation systems
• Computer vision and image processing
• Combining ML models with LLMs

Use Cases Deep Dive

• Healthcare: diagnostic assistants, medical record analysis
• Legal: contract analysis, legal research automation
• HR: resume screening, candidate matching
• Finance: fraud detection, financial analysis
• Retail, Education, Customer Service

Required Architecture Components

• Embeddings: text embedding and semantic search
• RAG: Retrieval-Augmented Generation pipeline
• Agents: autonomous agents with tools and memory
• Guardrails: safety mechanisms and content filtering
• Observability: monitoring, logging, and evaluation