Vibe writing maxims

02 May 2025

Shreyas Prakash headshot

Shreyas Prakash

Some vibe-writing maxims:

  • While writing, have two windows open: one for the writing, and the other one for ChatGPT. Previously I used to consult ChatGPT a couple of times for internet research, but now the role has transitioned to be a more conversational thought partner, helping you riff-raff on the idea for the essay. (Let’s say you’re writing an essay about tarrifs, and you want to understand what the general consensus is, so that you could provide an unique insight that can counter-position to the general consensus. It’s particularly helpful to provide this). Deep Research is also effective in productizing the ‘secondary research’ part of any essay where you have to crawl all over the internet to then derive the analysis from (Deep Research is basically a Mckinsey level research analyst at your doorstep helping you with this part of the process)
  • It also helps with alternative word seeking. Let’s say we want to find an alternate word to “outrage”, and you can give some references to ground them so that it generates a close cousin you’re expecting. This can be done on a word level, sentence level, or even on a paragraph level.
  • AI can help in fleshing out the essay with a meta-analysis, your first pass roughest-rough draft (even half-way through), you could probe ChatGPT to analyse if you’re getting to the core of the argument quickly? is it building momentum? is it a good opening, or there could be some alternate openings, closings?
  • In essays which involve more argument analysis, I also apply some of these questions inspired by Lex Fridman’s interview style. For eg:
QuestionPurpose
Can you steelman the case for …?Elicit strongest argument for..
Can you strawman the case for..?Address counterarguments for..
Can we break this down into first principles…Analyse from basic axioms
What happens if we take this to the extremes..Test robustness when it comes to the edge cases
  • The more we use AI-assisted writing, there is a big risk of the writing sounding like AI. The line needs to be tread carefully where you use AI for asking good questions, identify flaws, improve/edit content, not to actually put words in your mouth. I also use this quite extensively to remove the filler words such as dwelve, em-dashes and the usual AI-generated copypasta: `Identify and remove any generic or overused phrases that make the writing sound artificial. Replace common AI phrases with more original, specific language.

Subscribe to get future posts via email (or grab the RSS feed). 2-3 ideas every month across design and tech

Read more

  1. Home is where the feeling of home residesreflections
  2. Notes on doing most good, doing some good, and living goodphilanthropy
  3. Notes on Londontravel
  4. 18 key ideas from reading Henrik Karlsson this monthwriting
  5. Stop arguing, and start drawing circles togethermathematics
  6. The whole world is just the snake eating its own tailmental-models
  7. Life lessons and hot takes from my 30slifestyle
  8. Building a skill for coherent science illustrationsscience
  9. My agentic engineering workflow (step by step)agentic-coding
  10. Hammock driven developmentagentic-coding
  11. Peculiar ways number three fits into our funny little brainsmental-models
  12. AI sandwich as a defacto principle for anything agentic engineering relatedagentic-coding
  13. Authority in the guise of evidencecritical-rationalism
  14. Map is not the territoryphilosophy
  15. Self hypnosis as a manifestation ritualmeditation
  16. Hegelian dialectic for structured reasoning with AI agentsphilosophy
  17. How I prepare for tough negotiations nowadaysnegotiation
  18. When should we steelthread somethingproduct-development
  19. Breadboarding, shaping, slicing, and steelthreading solutions with AI agentsproduct
  20. Healthy conflict in teams have a tipping pointteam-building
  21. How I deslopify AI writingwriting
  22. How I started building softwares with AI agents being non technicalagentic-coding
  23. Read raw transcriptswriting
  24. Legible and illegible tasks in organisationsproduct
  25. L2 Fat marker sketchesdesign
  26. Writing as moats for humanswriting
  27. Beauty of second degree probesdecision-making
  28. Boundary objects as the new prototypesprototyping
  29. One way door decisionsproduct
  30. Finished softwares should existproduct
  31. How I periodically rank my rough draftsobsidian
  32. Flipping questions on its headinterviewing
  33. Vibe writing maximswriting
  34. How I blog with Obsidian, Cloudflare, AstroJS, Githubwriting
  35. How I build greenfield apps with AI-assisted codingagentic-coding
  36. We have been scammed by the Gaussian distribution clubmathematics
  37. Classify incentive problems into stag hunts, and prisoners dilemmasgame-theory
  38. I was wrong about optimal stoppingmathematics
  39. Thinking like a shipmental-models
  40. Hyperpersonalised N=1 learningeducation
  41. New mediums for humans to complement superintelligenceagentic-coding
  42. Maxims for AI assisted codingagentic-coding
  43. Virtual bookshelvesaesthetics
  44. It's computational and AI everythingagentic-coding
  45. Public gardens, secret routesdigital-garden
  46. Git way of learning to codeagentic-coding
  47. Style Transfer in AI writingagentic-coding
  48. Understanding codebases without using codeagentic-coding
  49. Vibe coding with Cursoragentic-coding
  50. Virtuoso Guide for Personal Memory Systemsmemory
  51. Writing in Future Pastwriting
  52. Publish Originally, Syndicate Elsewhereblogging
  53. Poetic License of Designdesign
  54. Idea in the shower, testing before breakfastsoftware
  55. Technology and regulation have a dance of ice and firetechnology
  56. How I ship "stuff"software
  57. Writing is thinkingwriting
  58. Song of Shapes, Words and Pathscreativity
  59. How do we absorb ideas better?knowledge
  60. Read writers who operatewriting
  61. Brew your ideas lazilyideas
  62. Trees, Branches, Twigs and Leaves — Mental Models for Writingwriting
  63. Compound Interest of Private Noteswriting
  64. Conceptual Compression for LLMsagentic-coding
  65. Meta-analysis for contradictory research findingsdigital-health
  66. Proof of workproduct
  67. Gauging previous work of new joinees to the teamleadership
  68. Task management for product managersproduct
  69. Beauty of Zettelswriting
  70. Stitching React and Rails togetheragentic-coding
  71. Exploring "smart connections" for note takingwriting
  72. Deploying Home Cooked Apps with Railssoftware
  73. Repetitive Copypromptingwriting
  74. Questions to ask every decadejournalling
  75. Balancing work, time and focusproductivity
  76. Hyperlinks are like cashew nutswriting
  77. Brand treatments, Design Systems, Vibesdesign
  78. How to spot human writing on the internetwriting
  79. Can a thought be an algorithm?product
  80. Opportunity Harvestingcareers
  81. Everything is a prioritisation problemproduct
  82. How I do product roastsproduct
  83. The Modern Startup Stacksoftware
  84. In-person vision transmissionproduct
  85. How might we help children invent for social good?social-design
  86. The meeting before the meetingmeetings
  87. Design that's so bad it's actually gooddesign
  88. Lessons learnt interview prepping for product rolesinterviewing
  89. Obsessing over personal websitessoftware
  90. English is the hot new programming languagesoftware
  91. Better way to think about conflictsrisk-management
  92. The role of taste in building productsdesign
  93. Dear enterprises, we're tired of your subscriptionssoftware
  94. Products need not be user centereddesign
  95. World's most ancient public health problemsoftware
  96. Pluginisation of Modern Softwaredesign
  97. Let's make every work 'strategic'consulting
  98. Making Nielsen's heuristics more digestibledesign
  99. Startups are a fertile ground for risk takingentrepreneurship
  100. Insights are not just a salad of factsdesign
  101. Minimum Lovable Productproduct
  102. Methods are lifejackets not straight jacketsmethodology
  103. How to arrive at on-brand colours?design
  104. Minto principle for writing memoswriting
  105. Importance of Whytask-management
  106. Quality Ideas Trump Executionsoftware
  107. Why I prefer indie softwareslifestyle
  108. Use code only if no code failscode
  109. Self Marketing
  110. Personal Observation Techniquesdesign
  111. Design is a confusing worddesign
  112. A Primer to Service Design Blueprintsdesign
  113. Rapid Journey Prototypingdesign
  114. Visualise detailed file structures on CLIcli
  115. Do's and Don'ts of User Researchdesign
  116. Design Manifestodesign
  117. Complex project management for productproducts
  118. How might we enable patients and caregivers to overcome preventable health conditions?digital-health
  119. Pedagogy of the Uncharted — What for, and Where to?education
  120. Future of Equity with Ludovick Petersinterviewing
  121. Future of Ageing with Mehdi Yacoubiinterviewing
  122. Future of Tacit knowledge with Celeste Volpiinterviewing
  123. Future of Mental Health with Kavya Raointerviewing
  124. Future of unschooling with Che Vanniinterviewing
  125. Future of Rural Innovation with Thabiso Blak Mashabainterviewing
  126. Future of work with Laetitia Vitaudinterviewing
  127. How might we prevent acquired infections in hospitals?digital-health
  128. The why to endure any howentrepreneurship
  129. Design education amidst social tribulationsdesign
  130. How might we assist deafblind runners to navigate?social-design