The Curious Engineer

All blogs / Software / engineering practice

seangoedecke.com RSS feed

seangoedecke.com

Sean Goedecke's personal blog

Software / engineering practiceAI / MLEngineering management / careers

’20’21’22’23’24’25’26
Posts
52 from 25 Dec 2020 to 7 Nov 2026
Posting
weekly (49 in the last year)
Kind
personal blog
Feed
https://www.seangoedecke.com/rss.xml

2026 47

  1. Working with designers
  2. A database is a store for global variables
  3. LLMs are not superhuman at reading and writing neural network weights
  4. Audience capture
  5. Reliably aligning LLMs is impossible in principle
  6. Fast and slow ways to improve your reputation
  7. Elegance and intelligence are anticorrelated
  8. You can just ask AI to do things
  9. Getting good at AI takes time and practice
  10. Don't anthropomorphize big tech companies
  11. Software is slow because of bad design, not slow languages
  12. Laypeople support technical fields for technology, not art
  13. What requirements don't you have?
  14. Writing is no longer proof of work
  15. AI is driving a boom in modestly successful software projects
  16. Software engineers don't get to be Luddites about AI
  17. Don't diagnose your opponents
  18. Software's centaur age may last decades
  19. How to read code
  20. Superpersuasion will look like bribery
  21. Shipping is the foundation
  22. Do not build the LLM torture factory
  23. Human-AI partnerships are for alignment, not capability
  24. Advice to a beginning software engineer
  25. You should all be asking way more questions
  26. System One models like Jev can train their own replacements
  27. Grit your teeth and ship it
  28. Two techniques for working with System One models
  29. Jev means structured output is interesting again
  30. Tell agents the why, not just the how
  31. Slow developer experience will bottleneck fast models
  32. AI is breaking our proxies for expertise
  33. Don't build tools for AI agents
  34. They really do think AI might kill everyone
  35. Why we should anthropomorphize AI agents
  36. Automatically detecting AI text in my browser
  37. Radical responsibility means treating people like tools
  38. How to protect yourself from workslop
  39. You have to beat the models at something
  40. Selling out
  41. You should never be angry at work
  42. Good writing is obvious, not original
  43. popular
  44. LLMs reward expertise
  45. Blog about things you don't understand yet
  46. Saying the obvious thing
  47. Software engineering may no longer be a lifetime career

2025 3

  1. Writing for AIs is a good way to reach more humans
  2. To get better at technical writing, lower your expectations
  3. Writing a tech blog people want to read

2024 1

  1. How I ship projects at big tech companies

2020 1

  1. Writing good technical explanations