Param Aggarwal (@paramaggarwal) header
Param Aggarwal (@paramaggarwal) profile picture

Member of Technical Staff at @different_ai. I obsess over how the product feels, and how fast the next version ships. Ex - Groww, Cult, Flipkart, Myntra.

Bengaluru, India
Joined January 2011
Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)Photo by Param Aggarwal (@paramaggarwal)
UI still matters. All those people who saw the terminal agents era and said UI is dead, you are wrong.
happy acquisition day to all those who celebrate @pierrecomputer -> @OpenAI @conductor_build -> @Meta @humanlayer_dev -> @factory the best part: all 6 of these companies deploy on @vercel !
1
134
This is why OpenAI is behind Anthropic. They give in to momentary friction and do not think about the long term. Claude offers simple 5x/10x plans. OpenAI offers 5x/20x/50x plans. They will not be able to sustain this and will attempt to revert back again.
Day 25/ We are doubling all usage limits for all Pro 100/200/500 tiers and Business Premium, effective today. We heard your kind feedback and won’t be reducing usage limits for Pro 200, so your Work and Codex usage will stay the same. We’re also going to continue offering extended limits for dots for now, and usage of your dot will still continue to be separate from your default usage. And we'll of course also keep shipping better models and other efficiency improvements that increase your effective usage over time.
3
82
Same story during their GPT-5 saga to introduce personalisation to their model. None of that ended up mattering - but they give in to customer back-lash in the heat of the moment.
13
Same story with their constant usage limit resets.
8
Opus 5.5 is such a good model - it launched in September.
33
Laptops evolved till just having a web browser on it was the only app that mattered. Phones will evolve till just having an AI app on it will be the only app that matters.
35
Agent skills is the new “I’m a published author”
54
So @threepointone and @trq212 have both quit - what comes next?
58
Param Aggarwal retweeted
1/ Today we're launching @different_ai , my third company, with $48M in seed funding led by @htaneja at @generalcatalyst, with participation from @ravi_lsvp at @lightspeedvp. Different is an AI marketing system that runs demand generation end-to-end for lean B2B marketing teams.
59
36
11
415
346,272
Launched - this is what I've been working on for the past three years!
Today, Different is out of stealth with $48M in seed funding led by @generalcatalyst, with participation from @lightspeedvp. Every marketer is building a world. We're here to help you make yours stand apart.
2
92
Everything that makes AOT JS possible also requires removing the "bad" parts of javascript. It's a win win situation.
I'm equally excited. Jarred knows a thing or two about the ecosystem compatibility. For `scriptc` it should be possible to get it almost to 100% but there are two things that pure AOT cannot do: 1. eval 2. Computed file location require (which is really also eval) You can get really far on (2) in many cases by compiling the whole reachable file set but the general case cannot be supported. IMO it is time to rip eval from JS/TS. RCE is too dangerous in the age of AI. A single uncontrolled read of `.constructor` and you are toast. bun --aot needs to ship a 100MB or so of dependencies into the AOT to support the RCE vector. Probably not worth it. For the few legit cases, apps should opt in and ship QuickJS or a quickly vibed JS interpreter themselves. Respectively, I think the right goal is full ecosystem compatibility where you have AI rewrite the minimal set of libraries that cannot be otherwise supported.
1
54
Do your best work does not actually talk about the work - but about figuring out what best looks like.
37
Reminds me of Trackmania Turbo
38
Param Aggarwal retweeted
Hang it in the Louvre
11
24
3
502
8,181
Replying to @Amul_Coop
@Amul_Coop I ordered 20 Amul Almondo from online shop, but only 16 came. Order ID OID1649318. Complaint number P26071446. I've emailed you as well.
1
77
@Shefali_Vj packing issue again... One box was supposed to contain 12 packs and only contained 8 - the box was properly sealed when delivered. There is filling below 4 of the boxes for total of 8 delivered.
19
This app remains free but the entertainment is top-notch.
Replying to @thsottiaux
Is this why your 28 days are full of nothingburgers
1
66
Claude and Grok are in the major league. ChatGPT and Gemini are in the minor league.
39
They tried clarifying but still not super clear.
Sorry for any confusion! Cloud sessions run on your Pro or Max plan, like the rest of Claude Code. This promo is an optional one-time credit that your cloud sessions spend first, before falling back onto your normal plan usage. We hope you enjoy the credits!
170
Claude has serious communication issues. What they mean here is you get extra credits beyond your included subscription usage, not that it is now a paid feature.
Cloud sessions are officially available and out of research preview! They let you keep Claude Code working, even when your laptop is closed. Existing subscribers get a one-time credit to try them: $100 on Pro, $250 on Max.
124