By leading your team with transparency, embracing fast failure, and building systems that encourage honest feedback, you not only unlock better games—you build better teams.
Read more 👉 lttr.ai/AsHux
#Prototyping #FailFast #Pressstartleadership
3
Rest in thrust, it’s a must - your fiery static fire test delivered critical data that strengthens the next New Glenn stack, turning today’s pad lessons into tomorrow’s orbital victories for Blue Origin.
“Space is hard.” - Elon Musk
Sometimes she explodes on the pad to remind us we’re not gods yet.
But damn if we won’t keep lighting the candle. 💥🚀
#NewGlenn #FailFast
2
1
6
192
May 27
I’ve tried a few plugins that use LLMs to clean up Hex-Rays pseudocode, but none of them really fit my workflow, so I started building my own.
My main focus is Windows kernel driver analysis, so this one is tuned for that use case.
Left: raw Hex-Rays output
Right: output from my plugin
Current features:
- Uses a WDK-header-based kernel API profile for functions, enums, structs, and macros
- Reconstructs common kernel patterns like LIST_ENTRY traversal, CONTAINING_RECORD, critical regions, and pool alloc/free
- Converts numeric literals into symbolic names for NTSTATUS, pool flags, pool tags, etc.
- Simplifies noisy Hex-Rays casts and pointer arithmetic into struct fields and kernel macros
- Renames cleanup, failfast, and error labels into meaningful control-flow targets
- Combines deterministic rules with LLM assistance instead of relying only on the model
If there’s interest, I’ll share more soon.
9
25
190
14,412
May 7
再次尝试，用小项目可以，用大项目大 feature，并不好，我突然想起我很久以前写的一条
x.com/vikingmute/status/2035…
codex 推理设置为高的时候兜底代码越来越多 为了不报错越写越多 使用这个提示词反而是让它这种情况越来越加剧了，其实failfast 也是很重要的 让问题尽快暴露出来
所以再次回到我的原则：把推理设置成中 并且明确要求它不要过度设计不要写过度防御性代码 新项目不要过度考虑兼容性
同时再次用上了之前别人推荐的 Claude 的system prompt，再发一遍，因为这些原则真的很棒。
May 7
这个 Codex 5.5 的 hack，非常有意思：
提示词是：“你对当前实现 100% 有信心吗？如果没有，请找出所有可能的漏洞，提出合适的修复方案，然后不断重复这个循环，直到你对新实现在事实上达到 100% 自信为止。”
作者说：Codex 5.5 是我见过唯一具备真正 self awareness 的模型，它绝不会轻易做出高自信的断言，除非它真正把所有事情都验证过。
但是对 Opus 4.7 完全无效，因为这个货多轮只会输出 “you're absolutely right” 这样的奉承话。
我当前就是用一个小型的项目来试试看，确实不错，之后可以用大项目试试看，看看是不是像作者说的这样。
6
13
159
23,711
May 2
Stop Press.
Know any failed entrepreneurs in the Delhi NCR region?
Not a typo. Not a consolation category.
A fast-growing Series A D2C beauty brand - cosmetics, skincare, haircare, makeup , is looking for a Head of Growth Marketing. Early 30s. Delhi-based. High ownership, high ambiguity, high upside.
The founder doesn't want someone who joined three companies in four years chasing salary hikes. He wants someone who has skin in the game literally and figuratively.
Someone who understands that equity is not a bonus, it's the point.
If a candidate doesn't value the ESOP story, they are the wrong candidate. Full stop.
The failed entrepreneur fits because they've already made the mental shift from monthly comfort to long-term wealth creation. That's the rarest thing to find in a salaried hire.
250 applications in 36 hours..on @Linkedin. Most won't make it past the first screening call. This comment exists to help the right 5 find their way in.
Tag someone who fits. Or forward this quietly to the one person you know who should be reading it.
Thanks @GautamGhosh
for amplifying it yesterday..i had to course correct🙏
#GrowthMarketing #D2C #Delhi #StartupHiring #FailFast
May 1
Know any #failed entrepreneurs in the NCR region? A fast-growing Series A company wants to hire a Growth Marketing Head.
Someone in their early 30s. Opportunities in D2C beauty brand: cosmetics, skincare, haircare, makeup.
Not for the faint-hearted, though!
Let me give you a hint -here's what the culture fit is defined
• If someone joins purely for a salary hike today, they will leave in six months for someone who gives a bigger hike.
• Target profile: people who work for wealth creation, not salary increment
• The ESOP pitch must land. If a candidate does not value the equity story, they are the wrong candidate.
2
3
440
Apr 30
THAT’S HOW U HANDLE IT.
Once upon a time, when someone talks sh_t. Steps up to u. Out of the blue. For no fu_kin reason.
U gotta drop dat sh_t.
Squash it.
#failfast (Right? You creepy old gay man; tomorrow!)
3
6,742
Consulting startups in the US and Europe, I saw one mistake: founders love the product, not the customer's problem. Same in consulting. SnapyCall changes that. Test an expert in 5 minutes.
snapycall.com
#startupadvice #consulting #failfast #snapycall #microconsulting
1
36
Apr 15
Entrepreneurship is about failing fast and learning from it. Don't waste 6 months on a bad idea. Embrace what scares you and do it now, before regret sets in. #Entrepreneurship #FailFast #Persistence
2
3
31
Apr 10
failfast, learnfaster fuels midnight mint chip runs & bad jokes guess the difference between chaos & genius
1
1
Apr 2
Crossed $2M ARR in jus 3 weeks 🔥
& on the growth side...
We killed the complete UGC market.
By generating 30M views
In the last 30 days, using AI - UGC.
Bruh, we cracked it!
Something which started as:
> 5 - 10k views a day
> Almost 0 shares
> Publishing 10 videos a day
Fun part: we were suggested to shut it down!
Today it’s:
> 1M views a day
> 25k shares/day
> Publishing 300 videos/day
But how?
> We figured out systems & automated the generation, edit & upload process.
> Researched & made the content more shareable
> Implemented templated, which worked (failfast >> perfect )
And in the next 30 days,
We’ll be getting 10M/day fr.
Up for the growth!
Less goo!
PS: Do you want a detailed article on it?
Apr 2
3 weeks ago, we launched Runable 2.0
Today we hit $2M in ARR, making @runable_hq one of the fastest-ever to do it!
We are just getting started...
5
1
22
1,112
Mar 23
On the humility of failure:
Sometimes the lesson isn't visible until years after the wound.
@francispedraza
#Entrepreneurship
#FounderStories
#FailForward
#failfast
1
3
57
Mar 21
小红书刷到的 感同身受 不知道大家有这个感觉没 codex 推理设置为高的时候兜底代码越来越多 为了不抱错越写越多 其实failfast 也是很重要的 让问题尽快暴露出来
所以我都把推理设置成中 并且明确要求它不要过度设计不要写过度防御性代码 新项目不要过度考虑兼容性
36
10
154
35,324
Mar 12
Coming soon: `dagger check --failfast`
Designed to shorten your agent's feedback loop: instead of waiting for an entire suite to finish, agents get the first failure immediately and can start fixing it.
dagger.io/changelog/#fail-fa…
2
5
1,581
Failed experiment: Tried a new onboarding flow. Conversion dropped 15%. Back to basics.
#BuildInPublic #FailFast
3
36
Fail fast. Improve fast. Launch fast.
Building something users can't ignore. 🚀
#BuildInPublic #StartupLife #FailFast
1
2
36
Feb 9
Failed at the task.
Ended up meeting the chairman.
Walked away with a better lesson than the outcome I wanted.
Failure doesn’t always block you.
Sometimes it redirects you.
#FailForward #ISRO #FailFast #Failbutlearn
2
70
Feb 8
100% 👍
In AI tools, fail fast beats slow regression (2026's big issue).
Use multi-pass execution fallbacks to iterate. Avoids data bloat traps trending now.
Faster to abundance!
#FailFast #aissiamxtlefirstone
3
87
Jan 20
Just landed in SF to join @fdotinc with our Norwegian unicorn cake biz. 🦄🍰
Immediately dropped our entire Go-to-Market strategy on the floor.
The 5-second rule applies to $8,000 items, right?
#FoundersInc #SFTech #FailFast #BuildingInPublic #startup
@daymakerguy
8
5
29
1,894
Success isn’t magic.
It’s literally a while loop.
Fail.
Debug.
Retry.
Repeat.
Most people quit before the loop converges.
#RealTalk #CodingJourney #DevLife #FailFast #KeepShipping #TwitterTech
1
85