The AI engineer's playbook: turning static models into dynamic agents

📘 Overview

A technical deep dive into turning static LLMs into dynamic agents that can use tools, search data, and reason. This workshop helps participants understand:

• AI engineering — LLMs and agents explained in clear, business-friendly terms
• LLM applications — RAG and semantic search as a primary case study
• Agentic shift — the techniques that transformed simple chatbots into autonomous agents
• LLM optimization — optimizing LLM inference costs

Key takeaway: A workshop designed to provide conceptual clarity for leaders and working code for engineers.

What you will learn

• Foundations — embeddings, similarity metrics
• Transformers — the engine behind foundation models
• Vector databases — long-term memory that stores data as embeddings
• RAG evolution — moving beyond simple vector search to Agentic RAG
• Reasoning models — Chain of Thought, and what powers models like DeepSeek-R1
• Tool calling — how models learn to interact with external systems, APIs, and datastores
• Agents — combining LLMs with tool calling and persistent memory
• The bigger picture — an overview of LLM training, fine-tuning, alignment, and inference

Who should attend

This session is for decision-makers and engineers looking to make informed choices in real-world AI projects. We’ll cover the trade-offs behind each choice:

• Local LLMs vs. cloud-based APIs
• Navigating the vector database landscape
• RAG options, including hybrid search
• Deterministic vs. agentic workflows
• Reasoning vs. non-reasoning models
• Balancing latency vs. throughput

✅ Prerequisites

You’ll get the most out of this session with a baseline understanding of:

• LLM basics — familiarity with tokens, context windows, and the difference between training and inference
• Technical setup — a laptop with an IDE and Python installed, and/or familiarity with Google Colab
• API access — an API key for an LLM provider (OpenAI, Anthropic, or Gemini)
• Optional — a virtual environment with the necessary libraries (e.g. openai) pip-installed ahead of time

About the Instructor

Arvind Devaraj works on LLM and AI Engineering at Juspay. His background spans GPU programming at NVIDIA and deep learning at Embibe (Reliance Jio). He holds a Master’s in Computer Science from the Indian Institute of Science (IISc) Bangalore, and secured AIR 7 in GATE Computer Science.

How to attend this workshop

This workshop is open to:
🎟️ Fifth Elephant community members — https://hasgeek.com/fifthelephant#memberships
🎟️ Ticket holders for The Fifth Elephant annual conference — https://hasgeek.com/fifthelephant/enterprise-ai-in-production-meetup#tickets

This workshop is open to 30 participants (in-person) & hybrid access for remote attendees. Seats for in-person participants will be available on first-come-first-served basis. 🎟️

Need more info?

☎️ Call: (91) 7676332020
📧 Email: info@hasgeek.com