Saurabh Bhatnagar

I just published “How I scaled Machine Learning to a Billion dollars: Strategy” medium.com/p/how-i-scaled…

@lennysan Solo dev vs faceless org

Even though every AI company is building their own version of OpenClaw (which is smart!), I haven't seen any of them get anywhere near the love and passion that OpenClaw inspires. There's something special about the OpenClaw experience that's hard to copy.

@code_star Yes but hard times make…

I’m not going to say frontier models with good harnesses can’t solve incredibly difficult problems. I will say many people who tried fine tuning circa 2022-2024 were just way to earlier. The base models at the time just weren’t good enough to meaningfully be improved. The methods for generating training data were terrible, using models as judges was not economic or reliable and paying for human annotations was too far out of reach. Fast forward to 2026 and many people have developed sophisticated and realistic evaluation environments which if you squint look a lot like what you want for RL. Cost of tokens is way down and the intelligence per token is much higher. Training infrastructure is much simpler to setup, algorithms and data are better. It’s more possible than it has ever been to adapt models to solve real world problems with a small team.

This should convince you that devs are a dead market Unless those devs are agents of course

You can hide these !commands in html comments so people don't see them when reading the skill. The command executes without the AI even knowing about it.

@michaelfreedman This is awesome

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

@_joe_berg_ put it well core issues: 2 incentive misalignments - you share your valuable outcomes data to a vendor, they use that to improve their product for nothing in exchange - vendor wants to participate in your value chain, but they have no leverage and commodification - outcomes are well formed –– there's a reason all but one of the AI customer service don't have pricing pages –– it's because as soon as they list them it'll start a pricing war for a commodified product

@TurnerNovak And New York

TIL there’s shady stuff going on in the Toronto accelerator scene

@Winterrose Should’ve asked for goat

im beef

Very data efficient to compress action space to latent skill space using an autoencoder Can apply to a lot of things besides robots

@TheSeaMouse Ugh yeah

Remember when "experts" told us synthetic data would lead to model collapse?

1/ Fully automonous + maritime defense AI Institutional Investor: "Could Be the Next Big Investment Opportunity in Venture Capital... "...

@bo_wangbo Congrats on the 👶

Computer on phone allows me code while watching the baby ;)

@cyantist @travisk He saw the church of ai before most

I'm excited about @travisk but I'm more stoked about Anthony. Many people haven't met him, but he's a real life Doc Brown and a wonderful human being. The story arc here is profound on many levels.

I take it back x.com/cgtwts/status/…

Every VC should stop tweeting and just clip Travis

@sytaylor It’s like cancer They found a way to attach themselves to ai

All of crypto is pivoting to virtual cards and it's so weird to me as a fintech guy. Like yeah. It's a virtual card?

Opus 3.6 is great at figuring out what the query should be But gpt-5.3/4 xhigh are great at optimizing it

@wolfejosh No sorry Only in the short term Bona fides x.com/analyticsaurab…

one big coming surprise in AI MORE jobs created for people. but with a twist the increase in jobs will be for human taskrabbits running physical world errands (with no context of why they're doing task)... for AI agents working for other humans and possibly other AI agents

@Winterrose Ha

the agentic era moat: how easy is it for agents to buy your service? vs how hard is it for an agent to replicate your service?