• Home
  • /
  • Next-Gen LLM Engineering: Build, Fine-Tune & Deploy Production-Ready AI Applications
Find a workshopPay feeRequest a callbackVerify a certificate
AI · Moderate · Live online

Next-Gen LLM Engineering: Build, Fine-Tune & Deploy Production-Ready AI Applications

Go beyond basic ChatGPT wrappers. Master QLoRA fine-tuning, enterprise RAG pipelines, and cloud deployment in 3 intensive days—and walk away with a portfolio-ready AI project.

★ 4.1 (31 ratings)25–27 Nov 20265:30PM IST3 Days (60-90 Minutes each Day)Certificate

About This Course

Move beyond using ready-made AI tools and learn how to customize, fine-tune, evaluate and deploy open-source Large Language Models (LLMs). This advanced, practical workshop introduces the complete LLM customization workflow in a focused three-day format: selecting a suitable model, preparing domain-specific datasets, applying parameter-efficient fine-tuning, evaluating and optimizing model outputs, and deploying a working AI application. Participants will work with modern open-source models and widely used development frameworks.

Aim

Designed for the next generation of AI innovators, this workshop aims to take participants beyond conventional prompt engineering into the world of Advanced LLM Engineering and Agentic AI Development. It focuses on the practical transformation of open-source Large Language Models into customized, high-performance AI applications using LoRA, QLoRA, custom data fine-tuning, RAG pipelines, model optimization and intelligent AI agents.

Workshop Objectives

  • Understand advanced LLM architectures and Generative AI workflows.
  • Prepare custom datasets for LLM fine-tuning.
  • Apply LoRA, QLoRA and PEFT techniques.
  • Explore model alignment, evaluation and optimization.
  • Develop Retrieval-Augmented Generation (RAG) applications.
  • Build intelligent AI agents using tool calling.
  • Gain practical exposure to LLM deployment and API integration.
Next: Day-by-day →

Workshop Structure

Day 1:Enterprise RAG Systems & Grounded AI Architecture

  • Understanding model architectures (Llama 3, Mistral, Qwen, Gemma) and parameter sizes (7B vs. 8B vs. 70B).
  • API-based models vs. Self-hosted open-weights models: cost, privacy, and latency trade-offs.
  • Extracting clean text from messy real-world PDFs, tables, and documents.
  • Fixed-size vs. Recursive vs. Semantic chunking—how chunk size and overlap directly impact answer accuracy.
  • How text becomes vectors using Hugging Face embedding models (BGE / E5).
  • Setting up and querying local and cloud Vector DBs (FAISS and ChromaDB).
  • Hybrid search (combining keyword search + semantic search) and Re-ranking.
  • Prompt engineering for strict grounding (“Answer only from context”) and evaluating RAG quality.
  •  Tools: Python | Google Colab | Hugging Face Transformers | Hugging Face Datasets | Pandas | PyTorch | JSONL
  •  Hands-On Activity: Smart Policy & Research Assistant — A working Python pipeline that ingests multi-page PDFs and answers complex queries with exact page citations.

Day 2: Hands-On LLM Fine-Tuning, QLoRA Optimization & Model Evaluation

  • When to use RAG (dynamic facts/knowledge) vs. Fine-Tuning (domain tone, structured JSON output, specialized reasoning).
  •  Dataset Engineering (The Secret to Good Models)
  • Preparing and cleaning instruction-tuning datasets (Alpaca and ChatML formats).
  • Using synthetic data generation to expand small datasets into high-quality training pairs.
  • Why full fine-tuning crashes consumer GPUs (VRAM math simplified).
  • How Low-Rank Adaptation (LoRA) freezes base weights and trains lightweight adapter matrices, combined with 4-bit NormalFloat (NF4) quantization.
  • Hands-on training using Hugging Face TRL (SFTTrainer), PEFT, bitsandbytes, and Unsloth (2x faster training with 60% less memory).
  • Monitoring training loss, preventing overfitting, and merging LoRA adapters with the base model.
  •  Tools: Google Colab | Python | Hugging Face Transformers | PEFT | TRL | Unsloth | bitsandbytes | PyTorch
  •  Hands-On Build: Domain-Specialized Custom LLM — Students fine-tune an open-source model on a niche dataset (e.g., Indian Legal Q&A, Medical Triage, or SQL Code Generation) and push their custom model weights to Hugging Face Hub

 Day 3:Inference Optimization, AI Agents & Full-Stack Deployment

  •  Model Compression & Fast Inference
  • Converting models to GGUF format and running them locally with Ollama.
  • Understanding inference engines (vLLM / llama.cpp) and quantization trade-offs (Q4_K_M vs. Q8_0).
  •  Intro to Tool-Using AI Agents
  • Giving your custom LLM access to external tools (web search, calculators, or custom Python functions) using LangChain / LlamaIndex.
  •  Building the Backend & Frontend
  • Wrapping the LLM pipeline into a clean REST API using FastAPI.
  • Building an interactive chat interface with streaming responses using Gradio or Streamlit.
  •  Cloud Deployment & Resume Packaging
  • Deploying the live application for free on Hugging Face Spaces or Render.
  • Structuring a recruiter-ready README.md, architecture diagram, and GitHub repository.
  •  Tools: Python | Hugging Face | FastAPI | FAISS | Google Colab | Ollama (Demo) | vLLM (Demo) | LangChain (Optional)
  • Day 3 Hands-On Build: Live Hosted GenAI Application — Every participant leaves with a public URL of their working AI app to add directly to their LinkedIn and resume.
Next: Dates →

Important Dates

Registration Ends

November 25, 2026
IST 4:30 PM IST

Workshop Dates

November 25, 2026 – November 27, 2026
IST 5:30PM

Workshop Outcomes

  • Understand modern open-source LLM architectures and model-selection considerations.
  • Choose between prompting, RAG and fine-tuning for different application needs.
  • Prepare and validate custom instruction datasets for model adaptation.
  • Apply LoRA or QLoRA in a guided supervised fine-tuning workflow.
  • Understand preference optimization concepts such as DPO.
  • Evaluate base and fine-tuned models using appropriate quality measures.
  • Understand quantization and inference optimization options.
  • Build a basic API and explore RAG or tool integration for a customized LLM.
Next: Fees →

Fee Structure

Student

₹2,999 | $65

Enrol as Student

Ph.D. Scholar / Researcher

₹3,999 | $75

Enrol as Ph.D. Scholar / Researcher

Academician / Faculty

₹4,999 | $95

Enrol as Academician / Faculty

Industry Professional

₹5,999 | $110

Enrol as Industry Professional

Next: FAQ →

Frequently asked questions

Will I get the recording?

Yes. Paid participants get the session recordings and study material in their NanoSchool dashboard (My registrations → Recorded Lectures / Study Material).

What if I miss a live session?

Watch the recording in your dashboard and send your questions to the mentor through the support desk.

Do I get a certificate?

Yes, an e-certificate of participation after the workshop, with a QR code anyone can verify and an "Add to LinkedIn" button.

Which fee category applies to me?

Choose the category that matches your current role: Student, Ph.D. Scholar / Researcher, Academician / Faculty or Industry Professional.

How can I pay?

By card, UPI or PayPal on the payment page; international bank transfer is also available there.

What is the refund policy?

See our Returns, Refunds & Cancellation policy: Returns, Refunds & Cancellation policy

Need Help?

We’re here for you!


(+91) 120-4781-217
WhatsApp +91 99581 61117

Registration open

More upcoming workshops

Live, mentor-led and certified. Dates and times in IST.

See all workshops
Support