teaching_llm_applications

Guest Lectures

Guest lectures bring practitioner and research perspectives that complement the core weekly material. Guests speak for approximately 45 minutes followed by 15 minutes of Q&A. Students are encouraged to read any suggested background material before each session.


Confirmed Speakers

Cole Robertson — Speech, Language, and LLMs in Industry

Background. Cole Robertson is the co-founder and CTO of a speech-AI startup building real-time spoken language understanding systems on top of large language models. His work sits at the intersection of automatic speech recognition (ASR), natural language understanding (NLU), and production LLM deployment.

Talk overview. Cole will discuss the engineering and product challenges of combining speech models with LLMs at scale. Topics will include:

Relevant course connections. Weeks 1 (LLM overview), 8 (prompting), 12 (production deployment).

Suggested background reading.


Glasgow Startup — LLMs for Real-World Applications

Background. This talk features the founding team of a Glasgow-based AI startup applying LLMs to a domain-specific business problem. Details to be confirmed; the talk will focus on the product development lifecycle from prototype to deployment.

Talk overview. The founders will walk through their experience building an LLM-powered product from scratch, covering:

Relevant course connections. Weeks 7 (fine-tuning), 8 (prompting), 9 (RAG), 11 (evaluation), 12 (applications).

Suggested background reading.


How to Prepare

Before each guest lecture, students should:

  1. Review the suggested background reading (listed above for each speaker).
  2. Look up the speaker’s prior work or company — understanding their context makes the Q&A more productive.
  3. Prepare at least one question in advance. Good question categories:
    • Technical depth: “How do you handle X in practice?” or “What did you try that didn’t work?”
    • Trade-offs: “Why did you choose X over Y?”
    • Failure modes: “What surprised you most when you deployed?”
    • Career/research path: “How did you move from research to product?” or “What skills do you wish you had earlier?”

Potential Future Speakers

The following areas are under consideration for future guest lectures. If you have a contact who works in one of these areas and would be willing to speak, please let the course organiser know.

Area Topics of interest
Healthcare AI Clinical NLP deployment, regulatory compliance, PPIE
AI Safety Red-teaming, alignment research, responsible scaling
Foundation model research Pre-training at scale, data curation, evaluation
Legal / policy AI regulation (EU AI Act, UK frameworks), IP and copyright
Open-source LLMs Community model development, GGUF/llama.cpp, Ollama
Multimodal systems Vision-language models, speech + vision

Notes for Speakers

If you are a guest speaker, thank you for contributing to this course. Practical tips: