Style Transfer in AI writing
Shreyas Prakash
The first draft usually generated by AI is so devoid of your style and personality, and to get to the stage where it even comes remotely close to that requires a lot of fine-tuning, and custom prompts.
Last week, I was looking at various ways in which I could “style transfer” my writing style to AI generated drafts. Due to the time consumed in style-transfer, I ended up writing on my own. (Read more here by watching this video essay)
And yet, I still feel that there is a way in which this could be more efficient. I used Claude Projects to get this working prototype in place, and here’s how I did this:

I knew that the best evergreen content for training would be my published blog posts. Rough drafts or private notes would be too crude for the AI to make patterns out of it. So I used this as a good benchmark to emulate my style. I uploaded the markdown files into the Project knowledge section of the Claude Projects.

You can upload relevant documents, text, code, or other files to a project’s knowledge base, which Claude will use to better understand the context and background for your individual chats within that project. In addition, you can define project instructions for each project to further tailor Claude’s responses. For example, instructing Claude to use a more formal tone or answer questions from the perspective of a specific role or industry.
I attempted to make a sample blog with this instruction:
I want to write essays that follow the style and flair of my previous blog posts. I don't want it to look like any AI-generated gibberish, copypasta. I want it to sound, feel and smell as if it's written by me: Shreyas Prakash.
Write a blog on 'moats for open source large language models'
While thinking through the possible answer for this prompt, Claude was able to analyse my language, patterns etc, to then adjust various aspects of the essay to better match my voice:
I've written this essay incorporating several elements I noticed in your writing style:
1. Use of vivid analogies and metaphors (Robin Hood, master chef's signature dish)
2. Breaking down complex topics into clear, digestible sections
3. Incorporation of real-world examples and companies
4. A conversational yet insightful tone
5. Forward-looking analysis that goes beyond the obvious
6. Personal perspective and observations
In the first version it produced, I noticed a slightly better draft than the generic ones usually generated.

I still felt like the headers and descriptions where making it look AI-ish, so I gave an additional prompt for it to flow well:

This was a good start. But it’s just that — a start. The next part of using this material before publishing is to reflect on the sentences, and to add my own spin to it.
To summarise, these are the steps I followed —
- Upload your blog markdown exports to the Claude projects knowledge base.
- Give a description to the project on Claude as : ‘Write new blogs on my personal website based on “style transfer” from the way I have written previous blogs’
- Provide a prompt to guide this process further —
I want to write essays that follow the style and flair of my previous blog posts. I don't want it to look like any AI-generated gibberish, copypasta. I want it to sound, feel and smell as if it's written by me: Shreyas Prakash. Write a blog on 'TOPIC' - Provide more guardrails such as — A strong opening hook, natural progression of ideas without explicit headers, characteristic use of concrete examples and metaphors, a more conversational tone throughout and with a signature email sign-off.
Subscribe to get future posts via email (or grab the RSS feed). 2-3 ideas every month across design and tech
2026
2025
- Legible and illegible tasks in organisations
- L2 Fat marker sketches
- Writing as moats for humans
- Beauty of second degree probes
- Read raw transcripts
- Boundary objects as the new prototypes
- One way door decisions
- Finished softwares should exist
- Essay Quality Ranker
- Export LLM conversations as snippets
- Flipping questions on its head
- Vibe writing maxims
- How I blog with Obsidian, Cloudflare, AstroJS, Github
- How I build greenfield apps with AI-assisted coding
- We have been scammed by the Gaussian distribution club
- Classify incentive problems into stag hunts, and prisoners dilemmas
- I was wrong about optimal stopping
- Thinking like a ship
- Hyperpersonalised N=1 learning
- New mediums for humans to complement superintelligence
- Maxims for AI assisted coding
- Personal Website Starter Kit
- Virtual bookshelves
- It's computational everything
- Public gardens, secret routes
- Git way of learning to code
- Kaomoji generator
- Style Transfer in AI writing
- Copy, Paste and Cite
- Understanding codebases without using code
- Vibe coding with Cursor
- Virtuoso Guide for Personal Memory Systems
- Writing in Future Past
- Publish Originally, Syndicate Elsewhere
- Poetic License of Design
- Idea in the shower, testing before breakfast
- Technology and regulation have a dance of ice and fire
- How I ship "stuff"
- Weekly TODO List on CLI
- Writing is thinking
- Song of Shapes, Words and Paths
- How do we absorb ideas better?
2024
- Read writers who operate
- Brew your ideas lazily
- Vibes
- Trees, Branches, Twigs and Leaves — Mental Models for Writing
- Compound Interest of Private Notes
- Conceptual Compression for LLMs
- Meta-analysis for contradictory research findings
- Beauty of Zettels
- Proof of work
- Gauging previous work of new joinees to the team
- Task management for product managers
- Stitching React and Rails together
- Exploring "smart connections" for note taking
- Deploying Home Cooked Apps with Rails
- Self Marketing
- Repetitive Copyprompting
- Questions to ask every decade
- Balancing work, time and focus
- Hyperlinks are like cashew nuts
- Brand treatments, Design Systems, Vibes
- How to spot human writing on the internet?
- Can a thought be an algorithm?
- Opportunity Harvesting
- How does AI affect UI?
- Everything is a prioritisation problem
- Now
- How I do product roasts
- The Modern Startup Stack
- In-person vision transmission
- How might we help children invent for social good?
- The meeting before the meeting
- Design that's so bad it's actually good
- Breaking the fourth wall of an interview
- Obsessing over personal websites
- Convert v0.dev React to Rails ViewComponents
- English is the hot new programming language
- Better way to think about conflicts
- The role of taste in building products
- World's most ancient public health problem
- Dear enterprises, we're tired of your subscriptions
- Products need not be user centered
- Pluginisation of Modern Software
- Let's make every work 'strategic'
- Making Nielsen's heuristics more digestible
- Startups are a fertile ground for risk taking
- Insights are not just a salad of facts
- Minimum Lovable Product
2023
- Methods are lifejackets not straight jackets
- How to arrive at on-brand colours?
- Minto principle for writing memos
- Importance of Why
- Quality Ideas Trump Execution
- How to hire a personal doctor
- Why I prefer indie softwares
- Use code only if no code fails
- Personal Observation Techniques
- Design is a confusing word
- A Primer to Service Design Blueprints
- Rapid Journey Prototyping
- Directory Structure Visualizer
- AI git commits
- Do's and Don'ts of User Research
- Design Manifesto
- Complex project management for product
2022
2020
- Future of Ageing with Mehdi Yacoubi
- Future of Equity with Ludovick Peters
- Future of Tacit knowledge with Celeste Volpi
- Future of Mental Health with Kavya Rao
- Future of Rural Innovation with Thabiso Blak Mashaba
- Future of unschooling with Che Vanni
- Future of work with Laetitia Vitaud
- How might we prevent acquired infections in hospitals?