sure why not
Hanover, NH
Joined April 2019
- Tweets178
- Following246
- Followers209
- Likes2.5K
Aamish Ahmad Beg retweeted
Teams like MongoDB, Workday, and Salesforce are building harnesses with Mastra that make skilled team members more efficient.
Today, we're launching the custom harness program, so you can co-build your own with our FDEs.
Let's chat: calendly.com/mastra-team/cus…
Aamish Ahmad Beg retweeted
Agent memory gets harder the longer a chat runs.
We wondered: how well does Observational Memory hold up at 100K, 1M and 10M tokens? So we ran it on BEAM, the long-term memory benchmark, with default settings.
Turns out it has the top published scores at 100K and 1M:
Aamish Ahmad Beg retweeted
Most LLMs get "context rot": recall decreases as context window approaches the context limit.
We wondered: do the new "System 1" decision models get context rot? So we tested Jev, Clef, and Luna Decisions across 400 questions.
Turns out decision models get context rot, too.
Aamish Ahmad Beg retweeted
“The man who can walk away from what he desires, for the sake of what he must become, is a man no one can own.”
- Epictetus
Aamish Ahmad Beg retweeted
BIG Update: I’m joining @a16z as an Investing Partner on the a16z @speedrun team! 🎉
I’ll be focusing on early-stage investing as part of speedrun, where we invest up to $1M in exceptional founders building great companies, & give them unfair advantages to succeed. If you are or know a founder who wants to win, I'd love to meet you! I'm especially excited about compelling novel ideas in AI dev tools, agent RL & evals, data, & infra in general.
For those who have known me over the years, you know how much joy I get in supporting those around me. From working countless nights to support my brilliant teammates at @scale_AI, to spending thousands of hours in my leisure time making free videos for millions of students in Bangladesh, I have given my all to support those doing their life's most important work. That has always been my personal mission.
So, when I met the speedrun team, a team that works relentlessly to make early bets & support founders from the very beginning of their journey, I knew this was the dream team for me. I am thrilled to join this shared mission with @andrewchen, @Tocelot, @tkexpress11, @emilybenn12, @kenanhsaleh, @far33d, @marcussegal, @ndrewlee, & the rest of the team! It's time to build.
Aamish Ahmad Beg retweeted
We’re proud to introduce GameEngineBench at
@NitrodeResearch, a benchmark for evaluating coding agents on real C++ runtime environments. Excited to share where we're headed as well.
Detailed 🧵 below.
Aamish Ahmad Beg retweeted
Last week a verified reporter messaged me about my startup. Here is how I caught her faking her whole identity to get to me.
Founders get messages like this all the time and most just say yes. Once you are on the call the person is really there to pull confidential details or to build a profile they can use later.
So I ran her through Sixtyfour's agents, and it came back saying she does not work at Bloomberg with 9 out of 10 confidence.
There was no author page, no bylines, and the name traced back to a real person in Florida whose identity was borrowed.
This is happening at scale and it isn't just fake reporters. Scammers, fraud rings, ban evaders, and counterfeit sellers can all be identified using agents now.
Platforms have the content signals. What they can't see is who is behind the account.
Our agents can.
Catch fakes before they catch you with Sixtyfour.
Aamish Ahmad Beg retweeted
We’re launching @trybaseframe to take you from AI fatigued to AI native.
You already have the AI tools. What you don't have is the time to figure out what to actually do with them, which features are useful to you, and how to use them.
BaseFrame does that for you. Its desktop app finds the repetitive work eating your week that you could automate today with the AI tools you already have.
Download the app and start unlocking hours today. Link below:
Aamish Ahmad Beg retweeted
Introducing HydraDB.
The graph native context infrastructure for agents. Purpose built to deliver precise context & observability into why agents act the way they do.
We've always believed graphs are the best way to manage AI context, but they've been too expensive to scale or impractical for storing full context. Until now.
@hydra_db combines in memory, NVMe, and object storage into a single graph layer, making context delivery faster, cheaper, and more precise.
We want context delivery to be extremely fast, 1000x cheap, and highly precise. Give your agents a brain.
Aamish Ahmad Beg retweeted
Day 2 of Sixtyfour Launches! Today, we're releasing our new people intelligence API!
Our deep people research sets a new state-of-the-art, outperforming Gemini 3.1 Pro, Grok 4.20, GPT 5.4, Exa Deep Reasoning, and Parallel Ultra while running 3x faster than the closest competitor.
Aamish Ahmad Beg retweeted
Agents should have access to every system. Introducing Integuru: generate fast, reliable APIs for any platform⚡️
API calls complete in ~3 seconds with 99.9%+ success.
No browsers or RPA. No data caching needed. With auth support.
@InteguruAI is free to use. Thread below.
Aamish Ahmad Beg retweeted
Introducing @sixtyfourai. AI Agents for people intelligence.
Since the invention of the internet, every person has been leaving breadcrumbs.
The child predator's username.
The scammer's banned accounts.
The people behind shell companies.
The prospect no contact database has.
The candidate's secret open-to-work signal.
The trail was always there.
Our agents have now made it possible to follow it.
Try it with the link in comments.
Aamish Ahmad Beg retweeted
A 6-person team is building task-specific AI models that are 4-8x faster than anything from OpenAI or Anthropic. 500K downloads on HuggingFace. No hype. Just better engineering winning on the merits.
This is what "make something people want" looks like in the model layer.
zeroentropy.dev
Aamish Ahmad Beg retweeted
We’re partnering with @lopus_ai to throw one of the biggest Launch Party on Saturday! 🎉
At Corgi, founders are at the heart of our community.
From founder dinners
to signature startup smoothies to our 24h Corgi Cafe, founders deserve great social events, so we make great events happen.
That’s why when we heard that Lopus will be hosting their Launch Party for 150+ founders this Saturday, we knew we had to be there to help out.
You will find us:
> Giving away smoothies
> Driving the Corgi Bus
> Interviewing founders
Location: Salesforce Park @ 2pm!
Or take the Corgi Bus from Corgi Cafe at 2pm :)
Aamish Ahmad Beg retweeted
Introducing Interact AI: a new interface for the web.
Add it to your website, and it talks to every visitor, answers questions, and shows your product.
Try it now -> interactlabs.ai
Aamish Ahmad Beg retweeted
A1Zap from @a1baseapi is the social network layer for AI apps. Starting with campuses, build apps and make interactive social content that spread through a unified social graph.
Congrats on the launch, @pashpops & @pennepastaa!
ycombinator.com/launches/Q24…
Aamish Ahmad Beg retweeted
Prompting is a design flaw.
So we built the thing they won't.
The entire AI industry convinced you that typing instructions into a box is the future. It's not. It's a crutch. You're doing the thinking, the context-setting, the remembering. The AI just autocompletes.
Felicity v3 is built to allow YOU to forget.
Forget Birthdays, Anniversaries, Coffee chats - Felicity will remember across everything:
The moment you wake up.
Felicity doesn't wait for prompts.
She prepares what you need, drafts your responses, and handles your to-dos before you think about them.
You don't prompt Felicity. You just confirm.
And she's on your phone, and life can be done without a laptop.
20x less overhead to get everything done right. Always one tap, swipe, or transcription away.
Entire interface rebuilt around voice (and text if you want).
Yesterday we shipped it. This isn't an upgrade. It's a different category.
Watch the demo.
heyfelicity.ai/
Tomorrow testflight goes for all.