Alexis Rondeau

Bookmarks

June 2025 · 70 bookmarks

Bookmarked from Alex Nguyen @alexcooldev

When I was building the MVP,
I spent 40 – 60 hours a week coding.

Now that it’s built
I spend 40 – 60 hours a week on marketing.

Building is just the beginning.
Getting people to care that’s the real grind.

Spend your time and effort on the right thing at the right stage.
We all get the same 24 hours a day.

Bookmarked from Тsфdiиg @tsoding

After all decades of wrestling with OOP patterns, Pure Functional Programming Bullshit and realizing that you just wanna write code - Dumb, Imperative, Straight to the point code - I feel scammed. I feel lied to. I wanna sue all the Best Practice preaching Shitty Book Selling mfs

Bookmarked from Andrej Karpathy @karpathy

+1 for "context engineering" over "prompt engineering".

People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window with just the right information for the next step. Science because doing this right involves task descriptions and explanations, few shot examples, RAG, related (possibly multimodal) data, tools, state and history, compacting... Too little or of the wrong form and the LLM doesn't have the right context for optimal performance. Too much or too irrelevant and the LLM costs might go up and performance might come down. Doing this well is highly non-trivial. And art because of the guiding intuition around LLM psychology of people spirits.

On top of context engineering itself, an LLM app has to:

  • break up problems just right into control flows
  • pack the context windows just right
  • dispatch calls to LLMs of the right kind and capability
  • handle generation-verification UIUX flows
  • a lot more - guardrails, security, evals, parallelism, prefetching, ...

So context engineering is just one small piece of an emerging thick layer of non-trivial software that coordinates individual LLM calls (and a lot more) into full LLM apps. The term "ChatGPT wrapper" is tired and really, really wrong.

Bookmarked from ludwig @ludwigABAP

they should never had given me access to Claude Code Max, I'm at 60M tokens inputted since I got access yesterday evening

I'm artisanally writing code while at work meanwhile Claude Code is cleaning up a bajillion various things I've accumulated, cleaning up my nix flake, improving my recent papers reimplementations (using the base paper + any code implementations it can find) etc...

Bookmarked from Manus @ManusAI

🚀 Introducing Manus Playbook

Each one is like a mini-app with clear guidance for any scenario:
📊 Business | 🎨 Creative | 📈 Sales & Marketing | 🎓 Education | 🎯 Fun & Life

Playbook is your step-by-step roadmap in Manus.

Bookmarked from Adam Lyttle @adamlyttleapps

This is the exact ASO strategy I use to consistently rank in top #10:

People tell me I'm doing ASO wrong… but it works for me (and only 1 of my apps got hit by recent App Store ranking changes...)

Judge for yourself:

Title field
"Primary Keyword: app name" to rank on both organic rankings and recommended searches

Example:
Fish Identifier: AI Scanner

Subtitle field
Give user and app store algorithm more context. Needs to be human readable but also have additional context for algorithm. Don't repeat primary keyword or any element of the keyword here

Example:
Species Recognition & Finder

Keyword field
Primary keyword exact match then rest of "keywords" simply context training for App Store algorithm

Example:
fish identifier,fish scanner,identification,scan,ai,recognition,....

Description field
Old-school SEO (repeat primary keyword 4-7 times, organically happens when you have keyword in app title... just repeat your title 4-7 times). Use chatgpt to turn your app (including features) into benefit driven description with your app name sprinkled through

What's New field
I make an effort to put my entire app title in the What's New field for another exact match keyword datapoint:

Example:
This version of Fish Identifier: AI Scanner includes usability updates and bug fixes

Bookmarked from Lenny Rachitsky @lennysan

Insane: 90-95% of Claude Code is now written by Claude Code.

I asked @mikeyk (Anthropic's CPO) what changes when so much of your code is written by AI?

His answer: The bottleneck moves from coding/design to:
1. Upstream: aligning the team around what to build.
2. Downstream: the PR merge queue, reviewing and merging all of the new code into production.

Anthropic is where other product teams will be in a year or two, so it's important we pay attention.

"I would expect that a year from now, the way that we are conceiving of building and shipping software changes a lot. Because it's gonna be very painful to do it the current way."

Bookmarked from Lee Robinson @leerob

things I believe:

  • shipping fast beats the best strategy

- speed is a superpower
- create a bias toward shipping
- small teams ship faster
- ai-native teams will move 10x faster than those not willing to change
- landings > launches (i.e. product adoption > shipping code)
- listen, build, ship, tell the customer, then repeat forever

  • you have no career ceiling

- grit > talent
- there’s no substitute for putting in the hours
- get 1% better every day

  • be ruthlessly truth seeking

- the truth can be painful
- you can just change your mind if wrong
- have strong opinions, loosely held
- maximize your exposure hours
- “anecdata” > data
- seek the collective truth, not just one opinion

  • communication is the job

- clear writing is clear thinking
- everyone (yes, you) needs to become a better writer
- leaders step up to provide clarity when absent
- be the person taking notes, even if it’s just for yourself
- mismatched expectations lead to sadness
- anticipate objections before hitting send, then address them

  • education is the best form of developer marketing

- be authentic and own your failures
- never use the word “webinar” ever again
- being helpful compounds

  • leadership means owning outcomes beyond the org chart

- influence > titles
- leaders have to do the work themselves and delegate
- you can write your own playbook
- study what worked for others, then take your own path

  • work can also be your hobby

- this doesn’t mean you can’t have other hobbies
- passion + boundaries > mythical “work-life balance”
- your best work comes from following your curiosity

  • demos > memos

- you could have built a prototype in @v0 during the meeting
- only ship things you’re excited about yourself

  • hiring is what separates good leaders from great

- there are two hiring answers: hell yes or no
- growth potential > current skill
- hire people you can learn from
- hire people you would someday be happy working for
- favorite interview question: what work are you most proud of?

  • always try to assume good intent

- lead with empathy
- they might just be having a bad day
- criticism is good feedback if you listen unemotionally

Bookmarked from Chris Bakke @ChrisJBakke

I feel fortunate that I moved to San Francisco 15 years ago and caught the very tail end of tech being weird and a bit nerdy.

Every startup in 2010 still had a few developers with ponytails, desks covered with hand-painted Dragon Quest figurines, a CTO who did 80 hours of community service in 2003 for installing malware on their high school computers, and someone in ops who had to leave the office at 6:15pm on Tuesdays to get down to Daly City for bagpipe lessons.

Now, everyone here is like "Oh man, I'm such a nerd! I'm super into hiking, rock climbing, and $400 mechanical keyboards for typing in Slack"

Bookmarked from Siro @Siron93

THIS APP IS A GIANT 🤯

"I am" makes $400K/month by tapping into the power of daily affirmations

Forget complex features

150K monthly downloads & a deeply loyal user base (1.5m followers on IG)

Let´s break it down 🧵👇
(1/17)