Large Language Models for Research

A workshop by: QCIF

https://www.qcif.edu.au/

Author: Moji Ghadimi

https://www.linkedin.com/in/moji-ghadimi/




🧠 Workshop Description:

Focus of this workshop

This workshop provides a practical, researcher-focused introduction to Large Language Models (LLMs), with a primary focus on:

  • Structured content summarisation
  • Code generation

Sectors this applies to

While the workshop is designed for academic users, the same techniques and code snippets can be easily repurposed across sectors:

  • Academia: creating structured literature reviews and research summaries
  • Industry: summarising technical or standards documents for engineering teams
  • Government: preparing policy summaries, briefs, and evidence-based guidance for decision-makers

What this workshop covers

  • How to use LLMs for summarisation, code generation, literature analysis, and knowledge synthesis
  • Foundations of prompt structure, retrieval grounding, avoiding hallucinations, and safe data handling
  • How to build transparent, auditable workflows suitable for academic, clinical, and technical domains
  • Understanding when LLMs help—and when they shouldn’t be used

Who this is for

  • Researchers interested in literature reviews or code generation
  • Anyone wanting structured, defensible LLM workflows


Setup Guide for Python and JupyterLab

This guide explains how to install Python, essential packages, and JupyterLab for running the LLM workshop notebooks on your computer or HPC environment.

📦 2. Install jupyterlab

Once Python is installed, you can install packages using pip (Python’s package manager).

Run this command in your terminal or command prompt:

pip install jupyterlab

  • jupyterlab is used for running notebooks interactively.

📦 3. Download notebooks

Download notebooks.zip from the left menu of workshop website and extract to your desired folder.

🧠 4. Launch JupyterLab

Open terminal or command prompt and navigate to the folder containing your workshop notebooks (or navigate then open a terminal), then run:

python -m jupyter lab

This will open a new tab in your default web browser with the JupyterLab interface.

You can then click on any notebook to open it.

⚙️ 5. Running on HPC

If you’re using a hosted JupyterLab instance on HPC infrastructure, Python and most dependencies are already installed. You typically only need to:

  1. Open the web-based JupyterLab interface.
  2. Load the correct Python environment (via module or dropdown).
  3. download and extract notebooks.zip from the left menu of the website.
  4. Upload notebooks to the HPC’s jupyter lab.
  5. Run 00_setup_guide.ipynb (this notebook).

📦 6. Install Required Python Packages

If the packages below are not installed uncomment (remove #) and run the cell below to install packages:

# remove # from the line below and run if packages are not installed 
# !pip install numpy pandas matplotlib openai requests

Note if you are running this in a terminal you need to remove “!” from teh start of the command. This will install:

  • numpy – for handling arrays.
  • pandas – for handling data.
  • matplotlib – for plotting.
  • openai – to connect to Groq or OpenAI-compatible APIs.
  • requests – for fetching data from APIs.

To verify installation, run the code below:

import importlib.metadata

packages = ["numpy", "pandas", "matplotlib", "openai", "requests"]
for pkg in packages:
    try:
        print(f"{pkg}: {importlib.metadata.version(pkg)}")
    except importlib.metadata.PackageNotFoundError:
        print(f"{pkg}: not installed")
numpy: 2.3.3
pandas: 2.3.3
matplotlib: 3.10.7
openai: 2.6.0
requests: 2.32.5

⚙️ 7: Get a Groq API key to access a llm online

  • Sign In to Your Groq Account. Go to https://console.groq.com/login.
  • If you already have an account, sign in.
  • If not, create a new account and complete the verification process.

Generate Your Groq API Key: 1. Navigate to the API Keys section from the left menu.
2. Click Create API Key. 3. In the pop-up window, enter a descriptive name for your key (e.g., AI Content Labs) in the Display name for the key field.
- This helps you easily identify the key later.
4. Click Submit.
5. Copy the displayed API key — it will only be shown once.
- Save it securely for future use.

✅ Summary

  • Install Python ≥ 3.10
  • Use pip to install required packages
  • Run jupyter lab
  • Verify everything by opening the example notebooks
  • Get a Groq API key