AI Accelerator Program

Building & EvaluatingAgentic AI Systems

Alessandro Romano

Senior Data Scientist

Kuehne+Nagel

Dipanjan Sarkar

Head of AI & Community

Analytics Vidhya

Raghav Bali

Principal Data Scientist

Delivery Hero
 Building & Evaluating Agentic AI Systems

About the Course

Video

introVideoByKJ

Building & Evaluating Agentic AI Systems is a practical, end-to-end program focused on designing, building, monitoring and evaluating Agentic AI systems that work reliably in real-world settings.

With live instructor-led sessions, 10 Guided Projects and 5 Mini Project Assignments, you will learn how to work with large language models (LLMs) and add in custom context with prompt engineering and retrieval-augmented generation (RAG). Next, you will focus on designing and building single-agent and multi-agent systems using industry standard frameworks and languages including Python, LangGraph & LangChain.

You will also learn standard design patterns including routing, tool-use, planning, reflection, multi-agent as well as other important components including context engineering, memory and model context protocol (MCP) to optimize and build effective Agentic AI systems.

Finally you will learn how to enable observability (AgentOps) focusing on evaluation, tracing and monitoring using methodologies and frameworks like LLM-as-a-Judge, agentic evals & LangSmith.

Prerequisites

You should be comfortable writing code in Python. Prior exposure to Generative AI is helpful but not required. What matters most is that you can code and understand core programming concepts.

Tools & Platforms

  • Python (Jupyter Notebooks using Google Colab)
  • LangChain, LangGraph
  • Gemini and OpenAI APIs (OpenAI keys would be provided for the duration of the course when needed)
  • Tavily Search API, Chroma DB, Ragas, DeepEval, Opik, LangSmith
  • ChatGPT, Gemini, Google AI Studio, Claude

Meet Your Instructors

Alessandro Romano

Alessandro Romano

Senior Data Scientist

Kuehne+Nagel

Alessandro Romano is a Senior Data Scientist, AI Engineer, and international speaker with a strong focus on building reliable, production-ready AI systems. With over 10 years of experience across mobility, logistics, and digital platforms, he has led and contributed to projects in dynamic pricing, demand forecasting, fraud detection, and large-scale experimentation systems. In recent years, Alessandro has specialized in agentic AI and LLM-based architectures, designing multi-agent systems for tasks such as automated code review, decision support, and intelligent workflow orchestration. His technical expertise spans Python, modern ML stacks, vector databases, and cloud-native deployment of AI services.

Dipanjan Sarkar

Dipanjan Sarkar

Head of AI & Community

Analytics Vidhya

Dipanjan Sarkar is currently the Head of Artificial Intelligence & Community, Analytics Vidhya. He is also a published Author, and Consultant, boasting over 13 years of extensive expertise in Machine Learning, Deep Learning, Generative AI, Agentic AI, Computer Vision, and Natural Language Processing. His leadership spans Fortune 100 enterprises to startups, crafting end-to-end AI products and pioneering AI upskilling programs. A seasoned advisor, Dipanjan advises a diverse clientele, from Engineers and Architects to C-suite executives and PhDs, across Advanced Analytics, AI Strategy & Development. Recognitions include 'Top 10 Data Scientists in India, 2020', '40 under 40 Data Scientists, 2021', 'Google Developer Expert - Machine Learning, 2019', 'Top 50 AI Thought Leaders, 2022', 'Google Champion Innovator Cloud AIML, 2022', 'Top 10 AI Leaders in India, 2024', '40 Under 40, 2025', alongside global accolades including Top 100 Influential AI Voices in LinkedIn.

Raghav Bali

Raghav Bali

Principal Data Scientist

Delivery Hero

Raghav Bali is a Principal Data Scientist at Delivery Hero, a leading food delivery service headquartered in Berlin, Germany. With 13+ years of expertise, he specializes in research and development of enterprise-level solutions leveraging Machine Learning, Deep Learning, Natural Language Processing, and Recommendation Engines for practical business applications. Besides his professional endeavors, Raghav is an esteemed mentor and an accomplished public speaker. He has contributed to multiple peer-reviewed papers and authored more than 8 books, including the second edition of his well received book Generative AI with Python and Pytorch. Additionally, he holds co-inventor credits on multiple patents in healthcare, machine learning, deep learning, and natural language processing.

Who Should Enroll

  • Data Scientists & Analysts
  • Machine Learning, Deep Learning & Generative AI practitioners
  • Data, Business & Technology Professionals
  • GenAI & Agentic AI enthusiasts
  • STEM Graduates
  • Data Professionals & Developers

    Seeking hands-on Agentic AI experience with LangGraph & LangChain

  • ML & GenAI Practitioners

    Strengthening portfolios with end-to-end projects

  • Business Professionals

    Applying GenAI & Agentic AI to real-world business problems

Key Features

Live Sessions

Live Sessions

5 hours/week

Projects

Projects

10 Guided Projects

Assignments

Assignments

5 Mini-Project Assignments

Cohort Experience

Cohort Experience

Discussion forum

Office Hours

Office Hours

Weekly discussions & QA

Recordings

Recordings

Lifetime access

Resources

Resources

All session materials

Certification

Certification

Course completion certificate

Support & Trust

Support & Trust

7-day money-back guarantee,

T&C

Course Curriculum

12 Live Sessions30 Learning Hours10 Guided Projects

Week 1

Essentials of Generative AI & Prompt Engineering

14 – 20 Feb 2026

Live Session

14 Feb · 2:30-5 PM UTC

Introduction to Generative AI & Agentic AI
  • Learn core Generative AI, LLM, RAG, and Agentic AI concepts
  • See real-world demos on popular LLM platforms
  • Get started with key frameworks and environment setup
Live Session

15 Feb · 2:30-5 PM UTC

Prompt Engineering & LangChain Essentials
  • Learn prompt engineering and LangChain basics
  • Apply advanced prompting techniques hands-on
  • Work with multimodal prompts across formats
  • Mini Project: Research & report generation using prompt engineering
Office Hours

18 Feb · 2:30-3:30 PM UTC

Week 1 Recap & Discussion on Mini Project 1 Solution (Optional)

Join in to discuss and ask any questions on the content or projects covered in Week 1. We will also release the solution code for Mini Project 1 Assignment and discuss it briefly during this time

Next cohort (Limited Seats)
14 Feb – 25 Mar 2026

Sold Out
$1199

7-day money-back guarantee

A Message from the Founder

The Vision Behind AI Accelerator Program

introVideoByKJ

Frequently Asked Questions

The AI Accelerator Program is a short, live, cohort-based, instructor-led program designed to upskill professionals in AI through hands-on learning. Within this program, we offer specialized tracks, such as Building and Evaluating Agentic AI Systems, where you'll focus on mastering specific AI skills. You'll learn live with a small group, interact directly with the instructor, and work on real-world projects rather than just watching pre-recorded videos.

This program lasts for 6 weeks, with 5 hours live sessions and 1 hour optional office hour per week, making it manageable alongside your regular job.

You will learn how to: 1. Work with LLMs using prompt engineering and RAG 2. Build single-agent and multi-agent systems using Python, LangChain, and LangGraph 3. Apply common agent patterns such as routing, tool use, planning, and reflection, along with context engineering, memory, and MCP 4. Add observability with a focus on evaluation, tracing, and monitoring, including approaches like LLM-as-a-judge and tooling such as LangSmith

The program is focused on building, monitoring, and evaluation. It is not positioned as a deep dive into production deployment and platform engineering (for example, full LLMOps pipelines, hosting, scaling, and enterprise deployment patterns).

Yes, the program includes 10 guided projects and 5 assignments. Each week, projects will be guided, and solutions for assignments will be provided during office hours. You’ll receive feedback, clarification, and support to ensure you stay on track throughout the program.

It’s fully live. All sessions are conducted in real time on Zoom, with fixed schedules and active participation expected.

This program is for people who want a hands-on, project-driven path to learn how to design and build RAG and agentic systems using Python and popular frameworks like LangChain and LangGraph.

This program is not a fit if you are not comfortable coding in Python (you do not need prior GenAI experience, but you do need solid programming basics). It is also not ideal if you are looking for a purely theoretical course with minimal hands-on work, since the program is built around live sessions, guided projects, and mini projects.

Yes. Hands-on work is designed around Python in Jupyter Notebooks using Google Colab. When paid APIs are required, OpenAI keys are provided for the duration of the course (when needed). The program also references common tools used in real builds (for example, vector DBs and evaluation tooling), so you get practice with an ecosystem you can reuse at work.

Yes! Upon successful completion of all assignments, you’ll receive a Certificate of Achievement. This certificate validates your hands-on expertise in applying AI concepts in practical settings.

Absolutely! This particular program is designed to equip you with the skills to build and evaluate Agentic AI systems, which you can immediately apply to your current projects or job responsibilities.

If you miss a live session, don’t worry! You’ll receive the recordings and resources used in the session as downloadable files, which you can keep and watch at your convenience. You’ll have 1 month to download the recordings and resources from the date the link is shared. However, we encourage you to attend live to maximize engagement and learning.

Yes, we offer a 7-day money-back guarantee. The 7-day period starts on Day 1 of your cohort and ends one day before the third live session. If you’re not satisfied with the program within this window, you can request a refund by contacting our support team at [email protected]. T&Cs.

While the schedule and curriculum are typically fixed, in exceptional circumstances, there may be adjustments. Any changes will be communicated in advance, and we’ll ensure that they don’t affect the core learning objectives. Our main goal is to provide you with a seamless, hands-on learning experience, and we’ll work to maintain that, no matter the adjustments.