home.social

#promptengineering — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #promptengineering, aggregated by home.social.

  1. 🚀 Fastest-growing AI projects today

    1. These tools not only growing rapidly but also accumulating significant interest as deve...
    2. The repository "ai-business-skills" by minhnv0807 a standout project with a Growth Scor...
    3. It production-ready skills and prompt templates for popular AI platforms like Claude Co...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  2. Fine-tuning vs. prompting — I ran the experiment so you don't have to. What actually determines which approach wins. hackernoon.com/ml-internals-fi #promptengineering

  3. Fine-tuning vs. prompting — I ran the experiment so you don't have to. What actually determines which approach wins. hackernoon.com/ml-internals-fi #promptengineering

  4. Fine-tuning vs. prompting — I ran the experiment so you don't have to. What actually determines which approach wins. hackernoon.com/ml-internals-fi #promptengineering

  5. Fine-tuning vs. prompting — I ran the experiment so you don't have to. What actually determines which approach wins. hackernoon.com/ml-internals-fi

  6. Fine-tuning vs. prompting — I ran the experiment so you don't have to. What actually determines which approach wins. hackernoon.com/ml-internals-fi #promptengineering

  7. Короткий промпт ≠ дешёвый промпт: как оптимизация ломает prefix cache в LLM-агентах

    32 tools в промпте - дешевле, чем 7. Да, да - если вы строите агентов, это не опечатка. Это следствие того, как работает prefix cache в агентском цикле, и почему локальная оптимизация одного запроса ломает кэш на всей траектории. Третья статья серии про prefix caching - теперь про этих ваших агентов.

    habr.com/ru/companies/bitrix/a

    #llmагент #prefix_caching #токены #aiагенты #ai #prompt_caching #promptengineering #contextengineering

  8. Короткий промпт ≠ дешёвый промпт: как оптимизация ломает prefix cache в LLM-агентах

    32 tools в промпте - дешевле, чем 7. Да, да - если вы строите агентов, это не опечатка. Это следствие того, как работает prefix cache в агентском цикле, и почему локальная оптимизация одного запроса ломает кэш на всей траектории. Третья статья серии про prefix caching - теперь про этих ваших агентов.

    habr.com/ru/companies/bitrix/a

    #llmагент #prefix_caching #токены #aiагенты #ai #prompt_caching #promptengineering #contextengineering

  9. Короткий промпт ≠ дешёвый промпт: как оптимизация ломает prefix cache в LLM-агентах

    32 tools в промпте - дешевле, чем 7. Да, да - если вы строите агентов, это не опечатка. Это следствие того, как работает prefix cache в агентском цикле, и почему локальная оптимизация одного запроса ломает кэш на всей траектории. Третья статья серии про prefix caching - теперь про этих ваших агентов.

    habr.com/ru/companies/bitrix/a

    #llmагент #prefix_caching #токены #aiагенты #ai #prompt_caching #promptengineering #contextengineering

  10. Короткий промпт ≠ дешёвый промпт: как оптимизация ломает prefix cache в LLM-агентах

    32 tools в промпте - дешевле, чем 7. Да, да - если вы строите агентов, это не опечатка. Это следствие того, как работает prefix cache в агентском цикле, и почему локальная оптимизация одного запроса ломает кэш на всей траектории. Третья статья серии про prefix caching - теперь про этих ваших агентов.

    habr.com/ru/companies/bitrix/a

    #llmагент #prefix_caching #токены #aiагенты #ai #prompt_caching #promptengineering #contextengineering

  11. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to fine-tune their prompts to get better resul...
    2. yaojingang/yao-open-prompts has seen significant traction with a Growth Score of 14.61...
    3. Its popularity can be attributed to the growing demand for high-quality prompts in non-...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  12. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to fine-tune their prompts to get better resul...
    2. yaojingang/yao-open-prompts has seen significant traction with a Growth Score of 14.61...
    3. Its popularity can be attributed to the growing demand for high-quality prompts in non-...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  13. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to fine-tune their prompts to get better resul...
    2. yaojingang/yao-open-prompts has seen significant traction with a Growth Score of 14.61...
    3. Its popularity can be attributed to the growing demand for high-quality prompts in non-...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  14. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to fine-tune their prompts to get better resul...
    2. yaojingang/yao-open-prompts has seen significant traction with a Growth Score of 14.61...
    3. Its popularity can be attributed to the growing demand for high-quality prompts in non-...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  15. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to fine-tune their prompts to get better resul...
    2. yaojingang/yao-open-prompts has seen significant traction with a Growth Score of 14.61...
    3. Its popularity can be attributed to the growing demand for high-quality prompts in non-...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  16. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to improve the efficiency and effectiveness of...
    2. yaojingang/yao-open-prompts has taken the top spot with a Growth Score of 16.62 and 1,3...
    3. Threpository a comprehensive library of Chinese AI prompts covering various scenarios s...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  17. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to improve the efficiency and effectiveness of...
    2. yaojingang/yao-open-prompts has taken the top spot with a Growth Score of 16.62 and 1,3...
    3. Threpository a comprehensive library of Chinese AI prompts covering various scenarios s...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  18. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to improve the efficiency and effectiveness of...
    2. yaojingang/yao-open-prompts has taken the top spot with a Growth Score of 16.62 and 1,3...
    3. Threpository a comprehensive library of Chinese AI prompts covering various scenarios s...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  19. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to improve the efficiency and effectiveness of...
    2. yaojingang/yao-open-prompts has taken the top spot with a Growth Score of 16.62 and 1,3...
    3. Threpository a comprehensive library of Chinese AI prompts covering various scenarios s...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  20. 🚀 Fastest-growing AI projects today

    1. Developers increasingly looking for ways to improve the efficiency and effectiveness of...
    2. yaojingang/yao-open-prompts has taken the top spot with a Growth Score of 16.62 and 1,3...
    3. Threpository a comprehensive library of Chinese AI prompts covering various scenarios s...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  21. 🚀 Fastest-growing AI projects today

    1. Developers seeking ways to refine their prompts to elicit sharper responses, while also...
    2. Our top growth scores reflect thtrend, with repositories showcasing innovative approach...
    3. Yao Open Prompts, with a Growth Score of 22.18 and 1,305 stars, a comprehensive library...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  22. 🚀 Fastest-growing AI projects today

    1. Developers seeking ways to refine their prompts to elicit sharper responses, while also...
    2. Our top growth scores reflect thtrend, with repositories showcasing innovative approach...
    3. Yao Open Prompts, with a Growth Score of 22.18 and 1,305 stars, a comprehensive library...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  23. 🚀 Fastest-growing AI projects today

    1. Developers seeking ways to refine their prompts to elicit sharper responses, while also...
    2. Our top growth scores reflect thtrend, with repositories showcasing innovative approach...
    3. Yao Open Prompts, with a Growth Score of 22.18 and 1,305 stars, a comprehensive library...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  24. 🚀 Fastest-growing AI projects today

    1. Developers seeking ways to refine their prompts to elicit sharper responses, while also...
    2. Our top growth scores reflect thtrend, with repositories showcasing innovative approach...
    3. Yao Open Prompts, with a Growth Score of 22.18 and 1,305 stars, a comprehensive library...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  25. 🚀 Fastest-growing AI projects today

    1. Developers seeking ways to refine their prompts to elicit sharper responses, while also...
    2. Our top growth scores reflect thtrend, with repositories showcasing innovative approach...
    3. Yao Open Prompts, with a Growth Score of 22.18 and 1,305 stars, a comprehensive library...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  26. 🚀 Fastest-growing AI projects today

    1. From prompt compression to multimodal injection testing, developers creating innovative...
    2. With growth scores ranging from 3 to 19, these projects gaining traction quickly.
    3. yaojingang/yao-open-prompts a comprehensive library of Chinese AI prompts covering vari...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  27. 🚀 Fastest-growing AI projects today

    1. From prompt compression to multimodal injection testing, developers creating innovative...
    2. With growth scores ranging from 3 to 19, these projects gaining traction quickly.
    3. yaojingang/yao-open-prompts a comprehensive library of Chinese AI prompts covering vari...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  28. 🚀 Fastest-growing AI projects today

    1. The trend shifting towards more specialized solutions that cater to specific use cases,...
    2. yaojingang/yao-open-prompts, with a growth score of 25.56 and 802 stars, a comprehensiv...
    3. Its popularity stems from its extensive coverage and relevance to the growing Chinese A...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  29. 🚀 Fastest-growing AI projects today

    1. The trend shifting towards more specialized solutions that cater to specific use cases,...
    2. yaojingang/yao-open-prompts, with a growth score of 25.56 and 802 stars, a comprehensiv...
    3. Its popularity stems from its extensive coverage and relevance to the growing Chinese A...

    Full report → pullrepo.com/report/todays-pro

    #AI #OpenSource #GitHub #Tech #PromptEngineering

  30. Early tests show that platforms can follow this format surprisingly well — especially when the tasks are clearly modularized.

    Read more 👉 lttr.ai/Aq2MV

    #ai #genai #promptengineering

  31. Early tests show that platforms can follow this format surprisingly well — especially when the tasks are clearly modularized.

    Read more 👉 lttr.ai/Aq2MV

    #ai #genai #promptengineering

  32. RE: mastodon.art/@otterlove/116505

    This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"

    Here are my suggestions—I thought, I didn't want to keep them from you:
    – How do you feel due to the fact that you are surrounded by #AI in your job and daily life?

    1/

    #amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM

  33. RE: mastodon.art/@otterlove/116505

    This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"

    Here are my suggestions—I thought, I didn't want to keep them from you:
    – How do you feel due to the fact that you are surrounded by #AI in your job and daily life?

    1/

    #amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM

  34. RE: mastodon.art/@otterlove/116505

    This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"

    Here are my suggestions—I thought, I didn't want to keep them from you:
    – How do you feel due to the fact that you are surrounded by #AI in your job and daily life?

    1/

    #amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM

  35. RE: mastodon.art/@otterlove/116505

    This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"

    Here are my suggestions—I thought, I didn't want to keep them from you:
    – How do you feel due to the fact that you are surrounded by #AI in your job and daily life?

    1/

    #amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM

  36. RE: mastodon.art/@otterlove/116505

    This year's #StateOfAI survey had the question "What other questions do you wish the survey would've asked, if any?"

    Here are my suggestions—I thought, I didn't want to keep them from you:
    – How do you feel due to the fact that you are surrounded by #AI in your job and daily life?

    1/

    #amCoding #amProgramming #artificalIntelligence #genAI #generativeAI #AGI #agenticAI #AIChatBot #vibeCoding #promptEngineering #agenticCoding #noAI #LLM

  37. MetaSPO meta-learns a task-agnostic system prompt via a bilevel loop: outer tunes system prompt across tasks, inner tunes per-task user prompts. Generalizes to 14 unseen tasks across 5 domains.

    The decomposition is the contribution. Once prompts split into task-agnostic (system) vs task-specific (user), meta-learning follows. System-prompt optimization transfers; user-prompt doesn't.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #KAIST #NeurIPS

  38. MetaSPO meta-learns a task-agnostic system prompt via a bilevel loop: outer tunes system prompt across tasks, inner tunes per-task user prompts. Generalizes to 14 unseen tasks across 5 domains.

    The decomposition is the contribution. Once prompts split into task-agnostic (system) vs task-specific (user), meta-learning follows. System-prompt optimization transfers; user-prompt doesn't.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #KAIST #NeurIPS

  39. MetaSPO meta-learns a task-agnostic system prompt via a bilevel loop: outer tunes system prompt across tasks, inner tunes per-task user prompts. Generalizes to 14 unseen tasks across 5 domains.

    The decomposition is the contribution. Once prompts split into task-agnostic (system) vs task-specific (user), meta-learning follows. System-prompt optimization transfers; user-prompt doesn't.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #KAIST #NeurIPS

  40. MetaSPO meta-learns a task-agnostic system prompt via a bilevel loop: outer tunes system prompt across tasks, inner tunes per-task user prompts. Generalizes to 14 unseen tasks across 5 domains.

    The decomposition is the contribution. Once prompts split into task-agnostic (system) vs task-specific (user), meta-learning follows. System-prompt optimization transfers; user-prompt doesn't.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #KAIST #NeurIPS

  41. MetaSPO meta-learns a task-agnostic system prompt via a bilevel loop: outer tunes system prompt across tasks, inner tunes per-task user prompts. Generalizes to 14 unseen tasks across 5 domains.

    The decomposition is the contribution. Once prompts split into task-agnostic (system) vs task-specific (user), meta-learning follows. System-prompt optimization transfers; user-prompt doesn't.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #KAIST #NeurIPS

  42. DSPy turns LM pipelines into typed-module graphs and compiles them end-to-end against a single metric, bootstrapping its own few-shot demonstrations.

    The programming-model layer is the real contribution, not any specific teleprompter. Once pipelines are typed graphs, pipeline-level search (MASS, MIPRO) becomes possible in a way it wasn't with string-template prompts.

    benjaminhan.net/posts/20260430

    #LLMs #AI #PromptEngineering #NLP #Stanford #ICLR