پن کیا گیا ٹویٹ
Pranav
264 posts
Pranav
@PranavAhl
Building a consumer health start-up. Prev building @haema_co, Prev buildspace sf2.
San Francisco, CA شامل ہوئے Aralık 2023
457 فالونگ310 فالوورز
Pranav ری ٹویٹ کیا
Do I take the safe role or bet on myself again?
As an immigrant, a job is permission to stay.
As a creative, I want to make films.
What would you do?
English
we achieved 37.6M views and 67k app downloads in february
we consolidated our data from over 6,000 videos to break down the biggest insights in tech ugc:
> the format that hit $1.48 CPI
> how to write content like you're texting a groupchat (and why it converts)
> when to post on IG instead of TT
& more
comment "PLAYBOOK" & i'll share the report with you
English
Pranav ری ٹویٹ کیا
I’m treating my job search like a growth problem, and I’ve got 45 days to crack it.
Instead of sending a resume, here's a 50-second video of my work.
If you’re looking to grow your product through storytelling and filmmaking, let's talk?
English
Pranav ری ٹویٹ کیا
Also celebrating this day :)
shai 🌻@shaiunterslak
happy 'missed another opportunity to create generational wealth' day to those who celebrate :)
English
Pranav ری ٹویٹ کیا
Every time I come on here there's another person announcing that they've broken ___ MRR or some opinion on how to run a company the right way
staying laser focused and filtering noise is one of the most important things when building
especially in this AI goldmine where there's so much value to be created
I've learnt this the hard way and still make mistakes 💀
Julia Pintar@juliapintar
this platform can easily become your worst enemy if you let it so much valuable information but at the cost of so much noise you need to become an expert at sifting through otherwise you will fall victim to the loss of focus (the worst thing that can happen to you)
English
For my app’s UGC program, we’ve been working on a ‘second pair of eyes’
Essentially an AI that can watch my creators videos and give feedback
It’s been super helpful to evaluate creators based on things that basic metrics couldn’t get.
Like ‘how often does creator x show feature y of my app in their videos’
Or like ‘what’s the ratio of experimental ideas vs tried and tested viral formats used by creator z’
Part of my effort to building a super efficient, well oiled machine
Nevin Ahluwalia@Nevinahl
The unofficial TikTok api on GitHub , is unbelievable for scraping data on TikTok Likes shares comments everything it’s on there and it’s implementation was key to the dashboard
English
Every-time I’m with my home friends they remind me of this 💀
I love the tech scene but it can feel overly opinionated and noisy sometimes
Important to take diverse perspectives on things
robyn@_robyn_smith
In sf, it’s important to have “tech friends” and “normal friends” Tech friends: - understand why you’re here - will inspire you to work harder Normal friends: - remind you the rest of the world doesn’t give a shit about any of the above - love a beer
English
acting is dead, never use real people again
how to make a viral studio-grade launch video with $150
no gatekeeping.
all the sauce on exact prompts, workflow, & execution.
[🧵DETAILED THREAD]
໊@christianyun
get roasted when you skip your habits MEET LYCHEE your AI pet that texts you in iMessage & emotionally manipulates you into productivity as your streak grows, you evolve your pet ALSO everyone who likes this post will own 0.0001% equity (yes, legally binding) for being a day 1
English
I’ve been grinding emails & DMs to influencers and this is exactly the experience I’ve had lol
Definately a numbers game though, you will find really great creators who will help grow your business if you keep hustling outreach
SingleClap@singleclap1
@_eth0n “Hi yes I’m interested my rates start at $100,000 per reel, but 1.5x if you want that cross posted to tiktok and another 3x for 30d of paid usage rights. If those prices aren’t high enough for you I can throw in a package deal and drop the price by 500 dollars, let me know!!!”
English
One of my biggest learnings of the year is to avoid scaling (hiring) too fast. My app saw over 100k users in it's first 3 months and immediately I invested most of the capital we earned on marketing and 'growing' the business.
But quickly found that I was spreading my time too thin and outsourcing things which I shouldn't have. Which lead to some slowdown.
Good thing its getting cold now, it's lock in time
I've made processes more lean and I'm working on building an insanely well oiled, efficient machine.
Biggest focuses rn are improving churn and hitting more viral videos consistently.
English
What does your personal site look like? Trying to get some inspo. Thanks!
English
AI memory was my first real engineering challenge for Entries
Its a journalling app that 'remembers' everything you write
It helps you reflect by asking questions and surfacing patterns in your thoughts, ideas and emotions.
Making the journal feel truly personalised had me immediately wary of using things like rag or storing occasional important 'memories'.
Such a system would have low resolution and in the case of a retrieval algorithm the response quality could be bottlenecked by the ability for the algorithm to find the correct things in the user's journal entries.
So similarly to what @rauchg is describing, I've built a system that utilises the large context window of modern models as much as possible and compresses journal entries as they get older.
IN addition to storing key memories about users.
So based on the user's input, every response as as personalised as possible.
Still madly tinkering with this system and building upon it, but clearly something's working right, I've scaled it to over 200k journallers and growing just this year!
Search myentries[dot]ai , always looking for feedback :)
Guillermo Rauch@rauchg
Single-threaded personal AI. Chatbots today assume that you’ll have multiple “conversations” and they’ll try to collect memories about you. In practice this is quite clunky. There might be an opportunity to break the mold, and maybe for a narrow usecase (like a truly personal assistant / confidant, rather than a general purpose search engine), have an app where it’s truly just one long chat. It’s all memories and context, that gets compressed in layers of memories over time to fit into the context (much like S3 storage tiers and CPU caches)
English