Name: Better RAG. Smarter RAG. - Advanced, Agentic RAGs
Start: 2025-12-06T14:00:00+05:30
End: 2025-12-06T17:30:00+05:30
Location: ThoughtWorks Technologies India Pvt Ltd. 4th Floor, Binarius Building, Deepak Complex, National Games Road

Topic: Better RAG. Smarter RAG.
Advanced, Agentic & Types of RAG — Demystified
About the speaker
https://www.linkedin.com/in/murari-ramuka-98a440a/

# ✅ **1\. RAG \(Retrieval\-Augmented Generation\) – Base Concept**

RAG = **LLM + external knowledge retrieval**
The model retrieves relevant chunks from a knowledge base and uses them to generate accurate, grounded answers.

***

# ✅ **2\. Advanced RAG – What It Means**

**Advanced RAG** refers to modern improvements over the basic RAG pipeline.
These are *techniques*, *optimizations*, and *enhancements* that make RAG more accurate, scalable, and production-ready.

### 🔹 **Key features of Advanced RAG**

1. **Better Retrieval**

* Hybrid search (vector + keyword)
* Multi-vector retrieval
* Reranking (cross-encoder)

2. **Better Chunking**

* Semantic chunking
* Recursive chunking
* Dynamic chunking depending on query

3. **Context Optimization**

* Context condensation
* Query rewriting
* Few-shot augmentation

4. **Advanced Orchestration**

* Multi-hop retrieval
* Tool-based retrieval

5. **Evaluation Frameworks**

* RAGAS
* TruLens
* Human + synthetic evals

📌 **Think of Advanced RAG as an optimized version of classical RAG.**

***

# ✅ **3\. Agentic RAG – Next Generation of RAG**

Agentic RAG introduces **agent-like behaviour**, where the LLM doesn’t just retrieve and answer—it **reasons, plans, decides, and iterates**.

### 🔹 **Key properties of Agentic RAG**

1. **Planning**

* Breaks the user query into steps
* Executes multi-step reasoning for retrieval

2. **Self-reflection**

* Evaluates its own retrieved results
* Refines instructions and retries

3. **Tool Use**

* Calls tools (database queries, APIs, calculators)

4. **Autonomous Actions**

* Runs loops: retrieve → evaluate → improve → answer

5. **Multi-Agent Systems**

* One agent for retrieval
* One for analysis
* One for verification

📌 **Agentic RAG = RAG + Autonomous Cognitive Loop + Tool Use**
The model becomes a *problem-solver*, not just a *retriever + generator*.

***

#

SUCHETA G DHERE

Pune Women in Machine Learning & Data Science

Technology

Artificial Intelligence

Big Data

Machine Learning

Data Visualization

Gender Equality

Data Science

Deep Learning

WiMLDS

Better RAG. Smarter RAG. - Advanced, Agentic RAGs

ThoughtWorks Technologies India Pvt Ltd. 4th Floor, Binarius Building, Deepak Complex, National Games Road

Share

Pune Women in Machine Learning & Data Science

Better RAG. Smarter RAG. - Advanced, Agentic RAGs

Pune Women in Machine Learning & Data Science

Details

Members are also interested in