BM25 Meets Embeddings: Build Hybrid Retrieval

Hosted by Aki Wijesundara and Manu Jayawardana

Tue, Jul 14, 2026

9:00 PM UTC (30 minutes)

Virtual (Zoom)

Free to join

Invite your network

Go deeper with a course

AI Engineering Bootcamp & Certificate
Aki Wijesundara, PhD and Manu Jayawardana
View syllabus

What you'll learn

Know what BM25 is good at

Understand why a classic keyword ranker still beats embeddings for exact words, names, and codes.

Fuse keyword and semantic scores

Combine the BM25 score and the embedding score into one ranking instead of choosing between them.

Tune the balance between them

Adjust how much each leg counts so retrieval leans toward precision or meaning as your data needs.

Why this topic matters

BM25 is the old reliable keyword ranker, embeddings are the meaning engine, and most people treat them as rivals when they are actually a great pair. In this free 30 minute session you build a retriever that scores documents with BM25 and with embeddings, then fuses the two scores so each covers the other's weak spots. You see exactly how the scores get combined and how that changes which document lands on top. You leave able to stand up a BM25 plus embeddings hybrid retriever and tune the balance between them.

You'll learn from

Aki Wijesundara

AI Advisor | Educator | Google AI Accelerator Alum

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.

Manu Jayawardana

AI Founder | Co-Founder & CEO at Krybe | Co-Founder of Snapdrum

Manu Jayawardana is a serial entrepreneur with multiple AI startup successes.

He is also the Co-Founder of Snapdrum, an AI consultancy helping companies integrate and automate their businesses using AI. Through Snapdrum, Manu has worked with top startups and enterprises across the U.S., Europe, and Asia, advising on automation, product development, and AI strategy.

Currently, Manu is the Co-Founder & CEO of Krybe, an AI voice agent startup in London transforming how businesses automate customer interactions. His experience spans fundraising, scaling products, and leading teams to deliver production-grade AI solutions adopted globally.

See all products from TAI

Sign up to join this lesson

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