Ruthuraraj R — AI Trainer & Generative AI Practitioner

Training Delivered

Workshops I've Organised & Led

Three hands-on workshops designed and delivered in 2026 — covering the full AI stack from ML fundamentals to production-ready agentic systems.

January 2026 · SNS College of Technology

3-Day Workshop

From Machine Learning to AI Agents: A Practical Hands-on Introduction

ML Fundamentals Deep Learning LLMs Prompt Engineering RAG AI Agents

Live demos of multi-agent workflows, LLM API integration, and AI-assisted applications. Covered Mini-RAG concepts and agent architectures for students and AI enthusiasts.

March 2026 · SNS College of Technology

1-Day Workshop

AI Content Studio: Multi-Agent AI System

CrewAI ChromaDB RAG Pipelines Multi-LLM Agent Patterns

CrewAI orchestration, ChromaDB-powered RAG workflows, and planner–worker–reviewer agent architectures for automated educational content generation.

May 2026 · SNS College of Technology

2-Day Workshop

Build Your Own Local AI Assistant: From Tools to Grounded AI Agents

Local LLMs Ollama RAG Tool Integration Grounded AI

Guided participants in building local AI assistant workflows from scratch — covering retrieval, tool use, and multi-agent systems with live hands-on demonstrations.

Portfolio

Projects & Experiments

Built from scratch to understand — not to benchmark. Every project documents what breaks, and why.

RAG-LMS · EdTech AI Flagship Project

R-B.A.T — RAG-Based Academic Tutor

A full Academic Console that transforms textbooks and PDFs into an intelligent, multi-mode learning system. Built with local LLMs, ChromaDB vector storage, semantic retrieval, and diagram-aware PDF grounding — designed to run entirely offline on student hardware.

Four distinct console modes — Tutor, Assessment, Presentation, and Evaluation — each powered by the same RAG backbone but with different prompting strategies and output formats. One knowledge base, four learning experiences.

Tutor Mode

Context-aware Q&A grounded in uploaded textbooks

Assessment Mode

Automated question bank generation from source material

Presentation Mode

Structured note & slide generation from PDF content

Evaluation Mode

Response scoring and feedback against retrieved context

Local LLMs ChromaDB SentenceTransformers pypdf Streamlit Ollama Diagram-Aware Retrieval

Architecture

PDF / Textbook Upload

↓ chunk · embed · index

ChromaDB Vector Store

↓ semantic retrieval

RAG Context Builder

↓ mode-specific prompting

🎓 Tutor

📝 Assessment

📊 Presentation

✅ Evaluation

↓ local LLM generation

Grounded Academic Response

RAG · Agentic AI

Workshop Assistant — ReAct-lite RAG Agent

Built as a live demo for the final session of the 3-day AI workshop. The agent answers questions grounded in workshop PDFs and notebooks using a heuristic ReAct-style decision loop.

The LLM routing call was replaced with a FAISS L2 distance threshold — eliminating one API call per query, making reasoning fully deterministic, and keeping agent logic transparent to a non-CS audience.

Gemini API SentenceTransformers FAISS pypdf Streamlit

GitHub →

PDF / Notebook Input

↓ chunk + embed

FAISS Vector Index

↓ L2 threshold routing

ReAct-lite Decision Loop

↓ grounded / general

Streamlit UI (labelled output)

Multi-Agent

AI Content Studio — Multi-Agent System

Transforms a topic prompt into structured educational content via four specialised agents: Research (RAG) → Image (diffusion) → Reviewer → Manager. Failure cases in agent handoffs and RAG retrieval are documented alongside the working pipeline.

CrewAI Gemini Groq FLUX RAG

GitHub →

Embeddings

Word2Vec Embedding Explorer

End-to-end Word2Vec (Skip-Gram + Negative Sampling) built from scratch in PyTorch, with an interactive browser-based explorer showing how semantic structure emerges from random vectors across training epochs.

Word2Vec PyTorch FAISS WikiText-2

GitHub → Live Demo →

Generative AI

Image Super-Resolution — SRGAN & ESRGAN

×4 super-resolution built from scratch with patch-based training, warm-up stability phases, and adversarial fine-tuning. Perceptual trade-offs and mode collapse behaviour are honestly documented without cherry-picked results.

SRGAN ESRGAN RRDB VGG Loss DIV2K

GitHub →

Multimodal AI

Image Caption Generator

Progresses from CNN–LSTM baseline (InceptionV3 + LSTM) through Transformer decoder to pretrained vision–language models. Validates learning via controlled overfitting and honestly documents why pretrained models outperform naive fine-tuning on small datasets.

InceptionV3 LSTM Transformer HuggingFace

GitHub →

Classical ML

XOR Problem — Why Deep Learning Exists

A concept-driven bridge from classical ML to deep learning. Deliberately breaks logistic regression on XOR before solving it with a single hidden layer MLP — focusing entirely on decision boundaries and architectural necessity, not accuracy.

scikit-learn MLP Decision Boundaries

GitHub →

Agentic AI · Multi-LLM

ReAct Web Research Agent

An autonomous research intelligence system built on the ReAct (Reasoning + Acting) paradigm. Given a topic, the agent generates 6 targeted research questions, then runs a full Thought → Action → Observation → Summary loop per question — combining Gemini for planning and synthesis with Groq LLaMA for fast, tight reasoning at each step.

The multi-LLM split is intentional: Gemini handles structure-heavy tasks (question generation, final report), while Groq LLaMA handles latency-sensitive reasoning loops. Two models, two roles — not interchangeable.

Planning

Gemini generates 6 structured research questions

Reasoning

Groq LLaMA generates Thought and Observation per cycle

Retrieval

Tavily Search fetches live web results per question

Output

Exportable report.md + full ReAct trace.md

Gemini Flash Lite Groq LLaMA 3.3 70B Tavily Search ReAct Loop Research Memory Streamlit

GitHub →

ReAct Loop (per question)

User Topic Input

↓ Gemini: generate 6 questions

Groq LLaMA: Generate Thought

↓ Action

Tavily: Live Web Search

↓ Observation

Groq LLaMA: Observation + Summary

↓ store in Research Memory

📄 report.md

🔍 trace.md

Agentic AI · LangGraph Latest Project

Reflective Travel Assistant

A memory-aware agentic travel planner built on LangGraph's StateGraph. The system collects weather and destination data, then runs a self-evaluation loop — the agent reflects on its own information quality, scores its confidence, identifies specific knowledge gaps, and conditionally re-searches before generating the final personalised itinerary.

The reflection loop produces observable, logged state transitions in the UI — "INSUFFICIENT (confidence 40%) → re-query → SUFFICIENT (confidence 80%)" — making the agent's internal reasoning auditable, not a black box.

Reflection Engine

LLM scores confidence + identifies missing topics per cycle

Conditional Re-search

Targeted Tavily queries for exactly the identified gaps

Image Enrichment

Pexels API attaches photos only after data quality confirmed

Personalisation

Diet, style, budget, transport enforced at prompt level

LangGraph LangChain Gemini Flash Lite Tavily Search WeatherAPI Pexels API Streamlit Session Memory

GitHub →

LangGraph StateGraph

WeatherAPI + Tavily Discovery

↓ initial collection

Reflection Engine (LLM)

↓ INSUFFICIENT? → re-search

Query Generator + Researcher

↑ loops back · max 2 cycles

Pexels Image Enrichment

↓ SUFFICIENT

Personalised Travel Guide

Ruthuraraj R.

Workshops I've Organised & Led

From Machine Learning to AI Agents: A Practical Hands-on Introduction

AI Content Studio: Multi-Agent AI System

Build Your Own Local AI Assistant: From Tools to Grounded AI Agents

Projects & Experiments

R-B.A.T — RAG-Based Academic Tutor

Workshop Assistant — ReAct-lite RAG Agent

AI Content Studio — Multi-Agent System

Word2Vec Embedding Explorer

Image Super-Resolution — SRGAN & ESRGAN

Image Caption Generator

XOR Problem — Why Deep Learning Exists

ReAct Web Research Agent

Reflective Travel Assistant

Skills & Tools

Certifications

Active on Coding Platforms

Let's Build Something Together