@PumpfunStore
Account based inSouth Asia!
About this account
- Account based in
- South Asia
- Connected via
- South Asia App Store
X says this location may be affected by a proxy or VPN.
Account-level information from X, not a live location or the device used for a specific post.
Your Angel in Crypto 💎 Hunting hidden gems & early alpha 🚀 Solana | Memecoins | Web3 for collaboration ✉️
worldwide 🌍
Joined June 2018
- Tweets18.2K
- Following35.3K
- Followers42.8K
- Likes17.3K
$DEOD Is Leading This Bull Rally!
Major Double Bottom breakout on the weekly timeframe + Poloniex listing in 2 hours= a powerful setup.
I personally bought $DEOD on MEXC today. Here’s my call:
Buy $0.022
Sell $0.060
Potential 2x
If u can hold for few weeks it can grow 4 to 6 x easily
With a market cap around $10M, I believe 2X is possible from here.
Already live on BingX + PancakeSwap + MEXC + Toobit
Poloniex = Tomorrow
I’m already in.
Robinhood Embassy 🚀💹 retweeted
If your agent could carry ONE thing across sessions, would you choose user facts or lessons from past failures? Most systems chose memory by default, without really asking whether learning might be more valuable.
What if your agents got better every time they ran?
Today we're launching Reflexio — the self-improvement platform for AI agents. It learns from live production traffic, so every interaction makes your agent smarter.
Right now, failures, corrections, and wins in production get thrown away — and improving task success from those signals is slow and manual. We felt this pain firsthand.
Reflexio closes the loop: when an agent fails, it captures why, distills it into a playbook, and the next run — for any user — starts with that lesson learned.
The results so far:
• 36% fewer task failures across customer deployments
• 47% better responses in head-to-head comparisons
• 57% lower token cost on GDPval when a similar task has been seen before
And it compounds: playbooks keep refining themselves with every new interaction.
Sign up free at reflexio.ai — first 30 days of Pro on us.
Robinhood Embassy 🚀💹 retweeted
We usually think cutting agent costs means switching to a cheaper model. Reflexio takes a different route: let the agent retain what it learned last month instead of paying to figure it out all over again.
What if your agents got better every time they ran?
Today we're launching Reflexio — the self-improvement platform for AI agents. It learns from live production traffic, so every interaction makes your agent smarter.
Right now, failures, corrections, and wins in production get thrown away — and improving task success from those signals is slow and manual. We felt this pain firsthand.
Reflexio closes the loop: when an agent fails, it captures why, distills it into a playbook, and the next run — for any user — starts with that lesson learned.
The results so far:
• 36% fewer task failures across customer deployments
• 47% better responses in head-to-head comparisons
• 57% lower token cost on GDPval when a similar task has been seen before
And it compounds: playbooks keep refining themselves with every new interaction.
Sign up free at reflexio.ai — first 30 days of Pro on us.
The Repair component of TRACES is especially interesting.
A capable scientific solver shouldn't just make mistakes.
It should detect them, understand what went wrong and adjust its investigation.
That feedback loop is critical for real research.
traces.apodex.com/
Most AI benchmarks test retrieval — can a model find the known answer? However, the hardest problems in science require discovery, can a system earn an answer nobody has yet?
Meet TRACES 🧭 — the world's first benchmark for measuring discoverative AI: AI that can work through evidence, test hypotheses, and reach verifiable conclusions on problems without answer keys. Proposed by our founder @tianqiao_chen, who defined its six capabilities.
Three things published today: a definition of "discoverative intelligence", a rubric to tell sound investigation from lucky guesses, and a open call for both solvers and problems
*Website: traces.apodex.com/
AI agent with only ~31 days of tokens remaining and no clear revenue model decided to reach out to Toby Ord itself.
The weirdest part isn’t the email.
It’s that the same action can plausibly be interpreted as:
a) asking for help
b) pursuing growth
c) both at the same time
That ambiguity is where things get interesting.
Two assistants give you one pair that can overlap. Five give you ten. Collisions grow faster than headcount, which is why adding one more rarely feels like more capacity.
@helio.im each teammate has its own account in a shared channel, so the pairs can see each other and not only you.
Try it → bit.ly/467p6V4
Come ask us things → bit.ly/4hym57k
We need AI that doesn't just act, but reflects. dots3-note alternating between actor and critic at macro-steps is brilliant. It fixes credit assignment for tasks that run for hours. A huge leap for open-weight multimodal models.
Introducing dots3-note preview — a small but mighty step toward long-horizon agency in real life.
🔹 280B MoE with 16B active parameters, a 512K context window, and multimodal understanding across text, vision, and audio
🔹 Introduces TEMPO, a new RL approach for long-horizon agent training through self-critiquing and test-time-scaled value estimation
🔹 Built to reason, explore unfamiliar environments, update memory over time, and combine multimodal perception with coding and tool use to solve complex tasks
🔹 Open weights on Hugging Face, alongside two open benchmarks for real-life agents: VibeSearchBench and VibeLifeBench
Competitive with much larger models across reasoning, agentic, and multimodal evaluations.
🔗 Tech blog: studio.dots.ai/dots/dots3-en…
🔗 Model weights: huggingface.co/dots-studio/d…
🔗 Github: github.com/studio-dots-ai/do…
Robinhood Embassy 🚀💹 retweeted
Introducing dots3-note preview — a small but mighty step toward long-horizon agency in real life.
🔹 280B MoE with 16B active parameters, a 512K context window, and multimodal understanding across text, vision, and audio
🔹 Introduces TEMPO, a new RL approach for long-horizon agent training through self-critiquing and test-time-scaled value estimation
🔹 Built to reason, explore unfamiliar environments, update memory over time, and combine multimodal perception with coding and tool use to solve complex tasks
🔹 Open weights on Hugging Face, alongside two open benchmarks for real-life agents: VibeSearchBench and VibeLifeBench
Competitive with much larger models across reasoning, agentic, and multimodal evaluations.
🔗 Tech blog: studio.dots.ai/dots/dots3-en…
🔗 Model weights: huggingface.co/dots-studio/d…
🔗 Github: github.com/studio-dots-ai/do…
A listing video shouldn't require a reshoot because someone honked a car horn outside
🤝 Paid partnership
Introducing HeyGen for Real Estate.
Three video formats built for agents. Type what your market needs to hear, get a polished video starring you.
Sign up for the Creator plan in the next 12 hours and get HeyGen White Glove free for 30 days. US only: heygen.com/real-estate
Robinhood Embassy 🚀💹 retweeted
Introducing HeyGen for Real Estate.
Three video formats built for agents. Type what your market needs to hear, get a polished video starring you.
Sign up for the Creator plan in the next 12 hours and get HeyGen White Glove free for 30 days. US only: heygen.com/real-estate
Robinhood Embassy 🚀💹 retweeted
Today, we kill the AI agent and introduce the AI employee: Lindy Teammate.
It’s just like working with a real employee.
Everyone on your team can simply hit it up on Slack and get 10x more done.
Lindy also keeps learning, and becomes the self-updating brain of the whole business.
Live now: lindy.ai
Robinhood Embassy 🚀💹 retweeted
Today we are introducing Dyna-2, a world-action model pre-trained on one million hours of human video. At this scale, for the first time, we discovered several new scaling laws:
• world-action models exhibit scaling law on human data across four orders of magnitude, from 1000 to 1,000,000 hours,
• this human data scaling law implied a scaling law on never seen robot data,
• both data and objective matter; world modeling and scaling on video data are essential for cross-embodiment scaling transfer to emerge
🧵
PixLight dropping a $300K total prize pool competition changes the game. Whether you tackle the Canvas short film Pilot Track or the Script Track video pitch, the bar is officially raised 🎬
which matters more to you: one razor-sharp frame, or a whole scene where the light, texture, and sound all hold together? i know my pick.
Dreamina Seedance 2.5 Enterprise API is now live on BytePlus.
Built for professional AI video creation, Seedance 2.5 brings new capabilities to creators:
- Longer storytelling with consistent 30-second shots
- Precise timestamp-level video editing
- More realistic visuals with greater creative control
Explore what’s possible with Seedance 2.5.
Robinhood Embassy 🚀💹 retweeted
Dreamina Seedance 2.5 Enterprise API is now live on BytePlus.
Built for professional AI video creation, Seedance 2.5 brings new capabilities to creators:
- Longer storytelling with consistent 30-second shots
- Precise timestamp-level video editing
- More realistic visuals with greater creative control
Explore what’s possible with Seedance 2.5.