Joined September 2024
11 Photos and videos
What happens once coding is (mostly) solved? Software engineering encapsulates so much work outside of the fairly menial task of coding, and with the launch of Factory 2.0, we are are shipping a suite of products that help engineers deliver the things you care about instead of having to manually shepherd each code change across the finish line. With your very own software factory, you can focus on building features that matter to your customers, make sure they actually work in production environments, and are well maintained after the features actually ship. Excited for everything were shipping here, more to come very soon!
Today, we're announcing Factory 2.0: from coding agents to software factories.
1
12
1,421
Shashank Sharma retweeted

29
76
653
188,336
Shashank Sharma retweeted
Introducing Automated Security Review in Droid.
5
17
175
30,027
Shashank Sharma retweeted

21
43
280
61,989
You can now have an open source model tell you to walk to the car wash instead of Opus while being 10x cheaper and 10x faster
Introducing model routing to Factory. Factory Router picks the right model for every task, automatically. Maintain frontier performance while cutting costs by 25%.
1
14
500
Shashank Sharma retweeted
bars
Replying to @FactoryAI
A higher token bill does not mean more work is getting done. One-line fix? Doc update? Too often, routine tasks get routed to the priciest path out of fear of losing performance. This only burns budget for no additional gain. You wouldn’t have Messi play goalie. Every model has different strengths whether it's reasoning, speed, cost, or context. Factory Router automatically picks the right model for every task.
1
16
1,066
Shashank Sharma retweeted
Introducing model routing to Factory. Factory Router picks the right model for every task, automatically. Maintain frontier performance while cutting costs by 25%.
199
215
2,206
2,158,820
Shashank Sharma retweeted
Update, in the last 9 days since launch this has saved our customers over a hundred billion tokens already. In addition to the context savings this provides, this helps prevent unnecessary context bloat which lets Droid stay focused on its current task for longer.
Introducing Deferred Context Engine in Droid.
9
13
266
42,108
Shashank Sharma retweeted
May 26
You know the drill 3 people will get 1 month of Droid Max (200$) plan Winners selected from comments: - share open source work you've done - say what you'd do with it - make me laugh Good luck, winners in 48 hours
94
8
147
18,498
RT @maddiehfaulkner: When @matanSF called me and said it's time to come build at @FactoryAI, I had to answer the call. Today, I'm announc…
2
13
Shashank Sharma retweeted
Every engineering org is trying to figure out their software factory right now. A lot of teams are building their own harnesses from scratch. Makes sense if you haven't talked to us tbh. But it's a lot of work, and it never stops. Our team is constantly thinking about every dimension of this problem. In this case, we were able to shave 40% off context size through smarter tool loading. A software factory's token volumes will be 10x that of an org with 100% human-agent adoption. Even a 5% efficiency gain is real money. The teams doing this at scale are already talking millions. To keep up when you're rolling your own harness, you have to keep doing that, forever, just to stay competitive on cost. And then you have to build your actual product. If your team is thinking about building their Software Factory, we have the building blocks ready to go. We'd love to chat!
Introducing Deferred Context Engine in Droid.
7
9
122
18,110
Shashank Sharma retweeted
Droid is now easier, faster, and more enjoyable to use in the Factory Desktop App. Excited to share the first thing I've been working on since joining @FactoryAI
8
6
48
14,358
Shashank Sharma retweeted

16
11
166
20,286
Jagged intelligence means models are good at different tasks! I have been primarily using Opus 4.7 for all my coding tasks, and then switching to GPT 5.2 for review. Try /review in droid, and take a look at our findings below:
Which model reviews code best? We benchmarked 13 models on AI code review across real PRs and the results are surprising. Spending more tokens did not result in better code review. A $1.25/PR model beat another that was more than 2x the cost. Meanwhile, budget models at $0.15/PR delivered ~80% of the quality of frontier models while being 10-30x cheaper. In fact, cost only explained ~21% of the difference in code review quality.
6
456
Shashank Sharma retweeted
The tests passed and the linters were green, yet the login form still broke in production. Why? Because nobody actually clicked through it. Today, we are launching Automated QA in Droids to fix this.
19
23
379
49,224
Shashank Sharma retweeted
Today we're opening access to Droid Computers: persistent machines for remotely orchestrating Droids. Spin one up in Factory's cloud or turn your machine into a Droid Computer. Either way, Droids have a dev environment with its own filesystem, credentials, and configurations.
30
49
713
223,519
I can't believe its been almost two years since I joined, but extremely proud of the Factory family and the product and company we've built together. This is just the start, can't stop won't stop until every developer in the entire world is using droids
Today, we are excited to announce our $150M Series C led by Khosla Ventures with strong participation from Sequoia Capital, Blackstone, Insight Partners, Evantic Capital, Abstract Ventures, 20VC, NEA, and Mantis VC. This puts our valuation at $1.5B and will accelerate our investment in research, product, and global go-to-market. Long live developers.
2
34
1,274