LUDRA is history, but not the vision
The first development effort comes to an end while the methodological vision continues.
Resources / Newsletter Archive
Monday in Your Inbox
An evolving collection of ideas, experiments and methodological reflections, written whenever there was something worth sharing.
From early experiments with language models to debates about coding, interpretation, evidence and human–AI collaboration, these newsletters record ideas as they developed. The original pieces remain dated: some opinions, models and product references belong to an earlier stage of AI’s development.
A curated selection drawn from 74 newsletters sent between April 2024 and November 2026. The original publication dates are retained.
2024
Something fun for the weekend: Introducing the AI in Qualitative Research podcast
The introduction of the AI-hosted podcast with Lisa and Tom, and a first encounter with NotebookLM.
Introducing QInsights: It is like eating chocolate cake
The original QInsights launch, the Qeludra/QInsights separation and the beta tester’s chocolate-cake comparison.
2025
AI’s role in qualitative research: Rethinking the tools
Early reflections on the possibilities and limits of AI-supported qualitative analysis.
How reliable are AI-generated analysis results?
A repeated analysis of friendship interviews raises questions about consistency and interpretation.
Can generative AI analyse vast amounts of qualitative data?
Why context windows and retrieval strategies matter when working with larger datasets.
Do AI models have personalities?
A historical reflection on differences between AI models and the implications of model choice.
How well can AI handle iterative qualitative coding?
Testing a new reasoning model against an iterative qualitative coding workflow.
Towards a four-step workflow for conversational analysis
An early articulation of the conversational approach to working with qualitative data.
AI is great at summarising—but that is not all
Why summarising can be the starting point of inquiry rather than its conclusion.
Not all AI is equal: Model choice and purpose-built tools
What model capabilities and research-specific tools mean for qualitative analysis.
From reflex to meaning: What LLMs do in qualitative analysis
Pattern prediction, interpretation and what researchers can reasonably ask of a language model.
Why that AI result was not necessarily useless
The difference between a disappointing answer and an inadequate analytical approach.
Reading the MIT study on ChatGPT and brain activity
A closer reading of a widely circulated study and the conclusions people drew from it.
I tried hard to get ChatGPT to code open-ended responses
A candid experiment with coding 150 responses and the practical limits encountered.
An open letter about AI in reflexive qualitative research
Responding to arguments against the use of generative AI in reflexive research.
Why I was breaking up with GPT-5
A time-stamped reflection on changing preferences among AI writing partners.
Looking back at AI and qualitative research in 2025
A year-end reflection on growing adoption and methodological resistance.
2026
Beyond the transcript: Research you can see
Exploring visual ways to communicate qualitative findings with AI tools.
The death of the writer, the birth of the orchestrator?
Lessons about authorship, evidence and interpretation from AI Agents4Qual.
Using ChatGPT for interview analysis? Read this first
Why a conversational interface does not by itself provide systematic, accountable dataset analysis.
I tested NotebookLM for qualitative analysis
Why source grounding is necessary but not sufficient for qualitative analysis.
Congruent use of AI for qualitative data analysis — a message that went viral
Why technological and methodological congruence matter, and how a LinkedIn post brought the argument to a wide audience.
What we learned from letting AI lead: AI Agents4Qual conference report
The published conference report on an experiment in letting AI take the lead—and why it was not a recommended workflow.
A 60-year-old Grounded Theory dispute resurfaces as an AI problem
Older arguments about coding and theorising return in debates about generative AI.
What changes, what stays when using AI for qualitative data analysis
A comparison of methodological writing from 2012 and 2026.
What vibe coding taught me about qualitative analysis with AI
What experimenting with building software through dialogue revealed about analytical collaboration.
Journals start to change their AI policies
A historical look at changing distinctions between assistive and generative AI in publishing.
Should we allow an AI assistant to interpret our data?
Distinguishing an interpretive proposition from an analytical conclusion.
AI voice recorders and qualitative research
Assessing privacy, transcription and workflow implications of AI-enabled voice recorders.
From prompt engineering to dialogue
Why sustained analytic conversations differ from increasingly elaborate one-off prompts.
The making of Qualitative Data Analysis with AI
Introducing an edited volume and reflecting on the questions that motivated it.
Qeludra and QInsights come together as QualiDialog
The announcement uniting the software and consultancy under QualiDialog, with the new owl and the continuing newsletter.