Master RAG From Chunking to Evals

Hosted by Karun Thankachan, Sai Kumar Bysani, and Dr. Aki Wijesundara

Sat, Aug 22, 2026

9:00 PM UTC (45 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

AI Engineering Bootcamp and Certificate: Build and Ship Production AI Agents
Dr. Aki Wijesundara and Manu Jayawardana
View syllabus

What you'll learn

Structure documents so retrieval actually works

Choose chunking and metadata that hold meaning together instead of splitting it apart.

Build the retrieval and ranking layer

Move past basic similarity search to hybrid retrieval and reranking that surfaces the right passage.

Measure retrieval quality with real numbers

Score retrieval directly, so you know the system is working before users tell you it is not

Diagnose the three ways RAG fails in production

Recognise chunking, retrieval and staleness failures from the symptoms users report.

Why this topic matters

Most teams have RAG running and quietly know it is not good enough. Answers are vague, citations are wrong, and nobody can say why. The instinct is to swap the model or rewrite the prompt, which rarely helps, because the failure is almost always in chunking, retrieval or the absence of any real measurement. Three engineers build the whole pipeline properly, in one session.

You'll learn from

Karun Thankachan

Senior Data Scientist, Walmart

Karun Thankachan is a Senior Data Scientist at Walmart E-Commerce, where he specializes in Recommender Systems, NLP, and Information Retrieval to improve item selection and availability at scale. He holds an MCDS degree from Carnegie Mellon University and has worked across E-Commerce, FinTech, PXT, and EdTech accumulating several published papers and 2 patents in Machine Learning.

Karun is the co-founder of BuildML, a community running weekly research paper discussions and monthly project development cohorts. He serves on the editorial boards of IJDKP and JDS, and is a recognized Data Science mentor on Topmate awarded Top 50 Creator in North America (2024) and Top 10 Data Mentor in the USA (2025). He also holds a Perplexity Business Fellowship and writes to 70k+ LinkedIn followers.

Sai Kumar Bysani

Lead Data Analyst, BlueCross BlueShield

Sai Kumar Bysani is a Lead Data Analyst at BlueCross BlueShield of South Carolina, where he applies machine learning and GenAI to healthcare data at scale. He holds an MS in Data Science from the University of Connecticut and holds 4 cloud certifications across AWS, Azure, and GCP with deep hands-on expertise in Python, TensorFlow, PyTorch, LangChain, and AWS SageMaker.

Sai is the founder of The Data Hustle, a Substack newsletter read across 41 US states, where he covers practical data science, AI career growth, and breaking into the industry. He writes to 168k+ LinkedIn followers, has directly impacted 10,000+ job seekers' career trajectories, and partners with HiCounselor to drive career acceleration in data and AI.

Dr. Aki Wijesundara

Co-Founder, Snapdrum · Senior AI Advisor, United Nations

Aki Wijesundara is an AI leader with a PhD in Machine Learning and extensive experience mentoring startups at Google’s AI Accelerator. With a career spanning both research and applied AI, Aki has taught 5,000+ students worldwide how to design and deploy production-ready AI systems.

He has worked across cutting-edge areas of applied AI, from LangChain and RAG pipelines to observability and large-scale deployment. As a researcher and educator, Aki bridges the gap between theory and practice, making complex systems approachable and actionable for engineers, founders, and product leaders.

Aki is also a frequent speaker and advisor to organizations adopting AI, helping them transition from experimentation to production at scale.


  • Ex–Google AI Accelerator researcher focused on responsible AI and applied ML.
  • PhD in AI & Cognitive Systems with published research across top universities.
  • Former researcher with teams affiliated with MIT, University of Oxford, & King’s College London.
  • Co-founder of Snapdrum delivered AI systems for finance, education, and healthcare.
  • Built and deployed AI product pipelines used by PMs, startups, and enterprise teams.
  • Instructor for multiple AI builder programs, helping 500+ professionals ship AI features fast.


See all products from TAI Labs

Sign up to join this lesson

By continuing, you agree to Maven's Terms and Privacy Policy.