Upskilling for the Agentic Era

Overview

This intensive session explores how to leverage Generative AI to streamline professional responsibilities. Participants will learn to move beyond basic prompts, focusing on high-ROI automation and creative augmentation across various business domains.

Hands-on Labs

• Document Summarization: Practice converting complex notes into executive briefs
• Writing Enhancement: Refine tone and clarity using generative techniques
• Task Automation: Build workflows to automate administrative processes

Module 01: The AI Paradigm Shift in Daily Operations

• Analyzing the transition to AI-augmented workplace efficiency
• Reviewing case studies of successful AI integration across industries
• Decoupling time-intensive tasks from manual labor via LLMs
• Finding the highest-impact "quick wins" for immediate AI adoption
• The trajectory of modern professional work in the era of large models
• Mapping workflows: Identifying candidates for full automation vs. co-pilot assistance
• Understanding the technical boundaries: Tokens, context windows, and latency

Technical Requirements

• Laptop with stable internet connection
• A free Google account
• A Zoom account for virtual delivery

Module 02: Precision Writing and Synthesis with LLMs

• High-stakes brainstorming for communication, strategy, and reporting
• Refining brand voice and clarity through advanced prompt engineering
• Semantic compression: Turning high-volume transcripts into actionable briefs
• Automated synthesis: Extracting key decisions and next steps from raw data
• Visual-textual synergy: Using AI to suggest and create supporting media
• The "Editor-in-Chief" approach: Using AI for recursive self-critique

Module 03: Creative Problem Solving with AI Sparring

• AI-facilitated brainstorming: Challenging assumptions and generating novel ideas
• Decoding complex business logic through structured AI frameworks
• Comparative analysis: Using AI to evaluate strategic alternatives
• Laboratory: Direct workflow mapping session for personal productivity
• Advanced Logic: Implementing Chain-of-Thought and Few-Shot prompting
• Navigating accuracy challenges: Tackling hallucination and logic errors
• The "Human-Centric" model: Building reliable oversight systems
• Security and data sovereignty in public and private LLM environments

Overview

Conducted by industry leaders in autonomic systems, this session focuses on the "Agentic" revolution—shifting from reactive chatbots to proactive, autonomous systems. Participants will architect and coordinate agent networks designed to execute complex business logic independently.

Hands-on Labs

• Agent Team Build: Design and deploy a multi-agent Crew for business scenarios
• Dual Tracks: Choose "Technical" (Python code) or "Strategic" (No-code UI) implementation
• Full Cycle: Define roles, test logic, and refine workflows for production

Module 01: Architecting the "Agentic" Workflow

• Deconstructing the Agentic framework: Memory, Autonomy, and Utility
• Chatbots vs. Agents: Analyzing the fundamental structural divide
• The logic of self-correction: How agents plan and execute multi-step paths
• ROI for Autonomy: Measuring value in agent-driven business processes
• Comparative Review: CrewAI, AutoGen, and the LangGraph ecosystem

Module 02: Designing Multi-Agent Crew Hierarchies

• Functional Personas: Assigning expertise and boundaries to specific agents
• Orchestration models: Manager-led vs. Sequential vs. Swarm architectures
• Defining clear boundaries: Guardrails for autonomous decision-making
• Blueprinting a "Cross-Functional Crew" for complex operational tasks
• The "Process Orchestrator": Coordinating communication and resource sharing

Module 03: Implementation & Safety

• Scalable Personas: Using YAML for standardized agent configuration
• Task Serialization: Building logic-rich instructions for agent execution
• Custom Utility: Connecting agents to private databases and enterprise APIs
• Observability: Analyzing agent logs to optimize communication paths
• Persistence Layer: Implementing Short-term and Long-term agent memory
• Validation Loops: Integrating "Human Feedback" into autonomous execution
• Multi-Modal Agents: Processing visual and structured data within a crew
• Performance Tuning: Managing token costs and rate-limiting at scale
• Defensive Design: Securing agentic systems against prompt-based attacks

Technical Requirements

• Technical Track: Python 3.10+, Jupyter Notebooks, basic understanding of LLM APIs (OpenAI/Anthropic)
• All: Laptop, stable internet, Zoom account

Overview

This expert-led program focuses on moving from brittle AI prototypes to resilient, enterprise-grade agentic systems. It deconstructs the shift from static prompt engineering to dynamic system design, centered on evaluation frameworks, reliable retrieval, and multi-agent coordination.

Hands-on Labs

• Core Agent Build: Develop a single-agent system with reasoning and memory
• Multi-Agent RAG: Orchestrate collaboration using Retrieval Augmented Generation
• Production Ops: Implement governance, semantic caching, and error handling
• Capstone: Design and defend a fully functional real-world agentic app

Module 01: System Design & Contextual Grounding

• Tracking the trajectory: From Deep Learning to Agentic Orchestration
• Architecting for Non-Determinism: Designing software that thrives on ambiguity
• High-Precision Retrieval: Context window optimization and reranking logic
• Hybrid Discovery: Combining semantic search with traditional indexing
• Grounding Principles: Ensuring LLM responses are anchored in fact

Module 02: Orchestration & Multi-Agent Swarms

• Autonomous Planning: How agents decompose complex goals into tasks
• Systems Architecture: Comparing Peer-to-Peer vs. Hierarchical agent models
• Communication Flow: Designing protocols for agent-to-agent data exchange
• Tool Standardization: Ephemeral & Persistent Memory for stateful sessions
• The MCP Standard: Future-proofing tool discoverability and integration

Module 03: Reliability, Scaling & AIOps

• Eval-Centric Development: Moving testing to the start of the build cycle
• Precision Benchmarking: Building "Ground Truth" sets for agentic accuracy
• Strategic Decisions: Using data to choose between Fine-tuning and RAG
• Production Monitoring: Red-teaming and safety layers for autonomous loops
• CI/CD for LLMs: Automated pipelines for prompt and weights deployment

Technical Requirements

• Basic coding knowledge (Python preferred)
• Laptop with stable internet
• API keys for major LLMs (OpenAI, Anthropic)

Overview

Most AI training focuses on "what" the technology is. This program focuses on "how" to wield it. We strip away the jargon to give leaders a clear framework for identifying high-value use cases, managing non-deterministic risks, and leading a workforce in transition.

Hands-on Strategy Labs

• The ROI Audit: A live, facilitated session calculating "Automation Arbitrage" potential. We quantify OpEx reduction and efficiency gains for specific workflows.
• Risk-First Prototyping: Paper-prototyping an AI feature with a specific focus on failure modes, brand safety, and compliance guardrails.
• The 90-Day Roadmap: Walking away with a concrete, budget-aligned execution plan. Includes simpler "Quick Wins" vs. long-term "moonshots".
• Vendor Evaluation Sprint: A mock RFP process where you learn to grill AI vendors on data usage, model provenance, and hidden costs.
• Crisis Simulation: A tabletop exercise managing a "runaway agent" or data leak scenario, testing your incident response playbook.

Module 01: The Generative Shift

• Probabilistic vs. Deterministic: Understanding why managing AI software requires a different leadership mindset.
• The Economics of Intelligence: Analyzing the falling cost of intelligence and how it reshapes business models.
• Beyond the Chatbot: Exploring "Agentic" workflows where AI takes action, not just answers questions.
• The Talent Gap: Identifying the new roles (Prompt Engineers, AI Ops) you need to hire or train for today.

Module 02: Identifying Value & Feasibility

• The "Co-Pilot" Matrix: A decision framework for choosing between "Human-in-the-loop" vs. "Fully Autonomous" systems.
• Data Readiness: Assessing if your proprietary data is actually ready for RAG. Strategies for cleaning unstructured data.
• Build vs. Buy vs. Fine-tune: Making the right infrastructure choices. Avoiding the "Wrapper Trap" and preventing vendor lock-in.

Module 03: Governing the Black Box

• Brand Safety & Hallucinations: Technical and operational strategies to prevent your AI from embarrassing the company.
• The "Human Firewall": Training your workforce to recognize social engineering and accidental data leakage.
• Regulatory Horizon: Preparing for the EU AI Act, NIST AI RMF, and future compliance requirements.
• Ethical Alignment: Establishing core principles to ensure your AI deployment matches your corporate values.

Module 04: Scaling & Operations

• LLMOps Overview: The infrastructure needed to monitor drift, cost, and latency at scale.
• Cost Management: Strategies for token optimization and model routing (using cheaper models for simpler tasks).
• Feedback Loops: Designing systems that get smarter with every user interaction.

Module 05: The Future Workforce

• Reskilling at Scale: How to transition your team from "doers" to "managers of agents".
• AI-Augmented Performance: Redefining KPIs and performance review in an AI-assisted world.
• Managing Culture Shock: Strategies for addressing employee anxiety and resistance to automation.

Overview

This intensive certification prepares product leaders to manage the transition from static features to non-deterministic, agentic systems. The curriculum emphasizes product intuition, feasibility analysis, and the deployment of autonomous workflows using modern low-code/no-code stacks.

Hands-on Labs

• Prototype Dev: Build/deploy 6+ AI-driven products using modern tools
• Strategic GTM: Create a Go-To-Market strategy and monetization model
• AI Discovery: Execute AI-enhanced product discovery activities

Module 01: Strategy & Feasibility Mapping

• Defining the expanded scope: PM responsibilities in the agentic cycle
• Agentic Fit: Identifying business units primed for autonomous intervention
• RAT (Riskiest Assumption Testing): Validating core AI hypotheses early
• Triangulating Value: Feasibility vs. Desirability vs. Operational Viability
• Market Analysis: Evaluating LLM providers and agentic frameworks

Module 02: Rapid Prototyping & Flow Design

• Low-Code Acceleration: Using v0.dev for instantaneous UI/UX mocks
• Automation Logic: Building autonomous paths via Crew.ai and n8n
• Persona Engineering: Creating system instructions that drive consistent behavior
• Boundary Logic: Setting operational constraints for agentic autonomy
• Semantic Integration: Connecting agents to legacy APIs and data silos

Module 03: Performance & Go-to-Market

• Intuition at Scale: Evaluating stochastic model outputs for product market fit
• Efficiency Metrics: Analyzing Cost, Latency, and Precision trade-offs
• Governance Models: Internal protocols for managing autonomous agent fleets
• GTM Strategy: Positioning and pricing autonomous AI products
• Resource Planning: Budgeting and team composition for AI-centric units
• Human-in-the-Loop (HITL) design patterns for user trust

Technical Requirements

• No heavy programming required; focus is on no-code/low-code tools (v0.dev, n8n, etc.)
• Laptop, internet, and Zoom account

Overview

This data-centric program replaces "vibes-based" AI development with rigorous Evaluation ("Evals") systems. Students will learn to construct systematic testing environments that bridge the gap between model potential and enterprise reliability.

Hands-on Labs

• Annotation Tools: Build interfaces for precise error analysis
• Test Suites: Design evaluation sets for edge cases and failures
• CI/CD Gates: Implement automated evaluation logic in pipelines
• RLHF Pipelines: Develop strategies for collecting human feedback

Module 01: The Lifecycle of the AI Evaluator

• Designing a unified eval lifecycle from prototype to production
• Beyond "Accuracy": Why traditional metrics fail for non-deterministic systems
• Systematic Failure Mapping: Identifying recurring points of model collapse
• Error Isolation: Tracing hallucinations back to prompt, logic, or retrieval errors
• Taxonomy of Failure: Categorizing and prioritizing system weaknesses

Module 02: Automated Judges & Dataset Scaling

• Strategic Automation: Aligning evaluators with high-level business objectives
• "Judges-as-a-Service": Best practices for using LLMs to evaluate other LLMs
• Information Quality: Prioritizing high-fidelity data over sheer volume in test sets
• Synthetic Hardening: Generating adversarial edge cases for robust testing
• Calibration: Testing the reliability and neutrality of automated eval systems

Module 03: Optimization, Monitoring & RAG Evals

• Quantitative Iteration: Refining prompt logic based on hard eval metrics
• Factuality Metrics: Measuring retrieval precision, grounding, and faithfulness
• Agentic Integrity: Evaluating logic flow within autonomous multi-step loops
• Infrastructure: Mastering tools like Arize, Langsmith, and Braintrust
• Live AIOps: Detecting semantic drift and regression in real-time
• CI/CD Guardrails: Embedding evals into deployment pipelines for safety

Technical Requirements

• Experience with AI/ML concepts is beneficial
• Familiarity with Python for building evaluation scripts
• Laptop, internet, and Zoom account