Praneeth Pike

on the occasion of launching windows & linux apps, we have a new promo video for rabbitholes[dot]ai 🔉 sound on

@shedrinkswater against marriage then?

@praneethpike i do not support forcing, god, no.

"Since I have learned to love you, I have closed my eyes for everyone else" ~ Rumi

@shedrinkswater normative relationships all are forcing to hold on to things permanently (until death)

@praneethpike wanting to hold things permanently is going against the laws of nature plus, it is something that can last until death does both of the people apart.

@zeeg ironically one of them has a neon sign that says “do the work”

Gotta love the grifters who just sit back and postulate on how the world works without living in it That podcast is best way to learn about wealth induced stupidity.

@brian_armstrong eastern mind: master inner world, suck at material world western mind: master outer world, suck at inner world

For me the best way to think about this: snorkel, don’t scuba. Diving too deep is a bit dangerous, and people get lost down there. Take glances at what’s beneath (lots of pretty fish!) from near the surface, and then get back to work. Wherever your motivation comes from (revenge, fear, competition, love of the game, learning, impact) is fine - harness it and get moving. I personally do some journaling (appreciations, write out goals, figure out what I want to get done tomorrow, etc) but then stop procrastinating and get after it.

it happened to me last year, and it happened to me this year. ai-first coding (I only participate in planning and stewardship) felt like a significant boost initially, but slowly bought me to a grinding halt. my shipping confidence tanked along with the reliability of the features I'm building. because my mind didn't come up with the plan, it becomes difficult for it to debug, abstract, or reliably plan the shipping cadence. I'm fighting my mind's training and specific advantage. I felt the fomo when tech influencers are on about agentic coding and their fancy setup with a million skills, agentic workflows, etc. I thought I should use it or else I'll be left behind. now that I zoom out, these people are not building products with true stakes – no users in production, no real money. their whole persona is about talking about what's trendy. it's tool calling, ai-sdk, effect, cursor, Claude Code, Ralph Wiggum, open claw, skills, tmux or whatever is next. no offense to them, but they're rewarded for sharing new knowledge and playing into the algorithm. a reminder to whoever needs it: don't let the algorithm capture your mind. if coding is your advantage, and you've been doing it for a while, you'll move faster by using ai just for autocomplete, as a pair programmer, review partner, or helping you cross uncharted territories. but we should not let it run the full show. we move fast when we keep our unique advantages

@shedrinkswater cause “anthropic” 😂

@praneethpike it feels human

are there any interesting things you came across or read related to giving a company a soul

@shedrinkswater what about anthropic’s branding??

@praneethpike omg yes yes, but like i was talking more in terms similar to anthropic's branding

I have a theory that node-based interfaces will win because from any screenshot you can immediately tell what’s going on, especially compared to linear prompting interfaces. This is extremely important for onboarding and adoption. The node-based stuff is largely a visual preference and kinda a gimmick (it’s no different functionally, at this point), but i’ve found it does make tutorials or client walkthroughs much easier

@michaelfreedman does work on windows?

Introducing TigerFS - a filesystem backed by PostgreSQL, and a filesystem interface to PostgreSQL. Idea is simple: Agents don't need fancy APIs or SDKs, they love the file system. ls, cat, find, grep. Pipelined UNIX tools. So let’s make files transactional and concurrent by backing them with a real database. There are two ways to use it: File-first: Write markdown, organize into directories. Writes are atomic, everything is auto-versioned. Any tool that works with files -- Claude Code, Cursor, grep, emacs -- just works. Multi-agent task coordination is just mv'ing files between todo/doing/done directories. Data-first: Mount any Postgres database and explore it with Unix tools. For large databases, chain filters into paths that push down to SQL: .by/customer_id/123/.order/created_at/.last/10/.export/json. Bulk import/export, no SQL needed, and ships with Claude Code skills. Every file is a real PostgreSQL row. Multiple agents and humans read and write concurrently with full ACID guarantees. The filesystem /is/ the API. Mounts via FUSE on Linux and NFS on macOS, no extra dependencies. Point it at an existing Postgres database, or spin up a free one on Tiger Cloud or Ghost. I built this mostly for agent workflows, but curious what else people would use it for. It's early but the core is solid. Feedback welcome. tigerfs.io

@GabriellaG439 yup. syntax is the most efficient format of communicating with computers, but it's not appealing to the current algorithm

New blog post: "A sufficiently detailed spec is code" I wrote this because I was tired of people claiming that the future of agentic coding is thoughtful specification work. As I show in the post, the reality devolves into slop pseudocode haskellforall.com/2026/03/a-suff…

@phuctm97 that's a perfect axiom. there's no argument

Happiness is a choice, not an outcome. Arguments are welcome.

@parmita sulking/moaning/overthinking ≠ introspection

Netscape guy: “yeah so when I was making the entirety of the internet, I didn’t typically sit around and think of how unfair high school was to me” Everyone:

@danielwchen0 @pmarca You’re absolutely right!

@praneethpike @pmarca can't introspect, can't think, can only spit out historical insights that are completely subjective and false. sounds like a broken llm 😭

Well.

@nbaschez @SoniaBaschez I feel you. virality on x is a function of extreme takes + existing reach + perceived social status

I have a sort of nuanced take on this which is that it's possible to be too introspective and possible to be not introspective enough and it's highly situation dependent Unfortunately does not make for a good viral tweet haha

@danielwchen0 @pmarca logic is thinking I guess 😭

@praneethpike @pmarca I know 😭 Does the absence of introspection lead to a complete breakdown of logic?

@zachpogrob you’re conflating introspection with overthinking

INTROSPECTION IS A LIE