Pinaki Dasgupta ,MBA ✨ @Seanku

Mgmt.Consultant/Business Exec. w/ interests in Science,Tech,IT & Fintech. #ML,#AI, #Cybersec #Robotics,#Cloudcomputing #IoT, #BigData,#DevOps, #Startups Worldwide Joined January 2010

Tweets

13K
Followers

709
Following

378
Likes

399

Pinaki Dasgupta ,MBA ✨ @Seanku

12 hours ago

PromptCoT 2.0 👇

elvis @omarsar0

a day ago

PromptCoT 2.0 👇

19 61 324 55K 315

Download Image

0 0 0 9 0

Pinaki Dasgupta ,MBA ✨ @Seanku

12 hours ago

Popular Tinycolor #npm Package Compromised in Supply Chain Attack Affecting 40+ Packages socket.dev/blog/tinycolor… by @SocketSecurity #supplychainattack

0 0 0 3 0

Pinaki Dasgupta ,MBA ✨ @Seanku

13 hours ago

Prompts as Code & Embedded Keys | The Hunt for #LLM -Enabled #Malware.

Tom Hegel @TomHegel

a week ago

Prompts as Code & Embedded Keys | The Hunt for #LLM -Enabled #Malware.

1 7 14 2K 2

Download Image

0 0 0 20 0

Anything @anythingai

3 days ago

Introducing Anything Max. Autonomous software engineer. Tries your app like a person. Solves 97%+ of the hardest bugs. Ships features on its own. Live now, reply for free weekend access

793 186 2K 2.1M 1K

Download Video

Pinaki Dasgupta ,MBA ✨ @Seanku

a week ago

DeepSeek R1 :This AI model teaches itself to think through problems, no humans required. nature.com/articles/s4158…

0 0 1 26 0

Download Image

Pinaki Dasgupta ,MBA ✨ @Seanku

a week ago

🚀SFR-DR-20B achieves 28.7% on Humanity's Last Exam (text-only) using only web search 📷, browsing 📷, surpassing DeepResearch w/ OpenAI o3 & Kimi Researcher. 🐍

Caiming Xiong @CaimingXiong

3 weeks ago

🚀SFR-DR-20B achieves 28.7% on Humanity's Last Exam (text-only) using only web search 📷, browsing 📷, surpassing DeepResearch w/ OpenAI o3 & Kimi Researcher. 🐍

28 146 958 305K 712

Download Image

0 0 0 80 0

Pinaki Dasgupta ,MBA ✨ @Seanku

a week ago

DeepTRACE : Auditing Deep Research AI Systems for Tracking Reliability Across Citations & Evidence.About 1/3rd of the statements made by AI tools were not supported by the sources they provided. @SFResearch arxiv.org/abs/2509.04499

0 0 0 36 0

Download Image

Pinaki Dasgupta ,MBA ✨ @Seanku

2 weeks ago

Illusions of AI consciousness : The belief that AI is conscious is not without risk. science.org/doi/10.1126/sc…

0 0 0 17 0

Shruti Codes @Shruti_0810

2 weeks ago

All Handwritten colorful notes for Free! Hurry up! Free for 24 hours. ✅ Python ✅ Node.js ✅ React.js ✅ JavaScript ✅ Oops in C++ ✅ HR interview ✅ Machine learning ✅ Cloud Computing ✅ Computer networks To Get it: 1. Follow me (so that i can DM you 2. Repost 3. Comment…

293 210 413 36K 216

Download Image

MIT CSAIL @MIT_CSAIL

2 weeks ago

Beginner-friendly LLM course exploring foundations, architectures, training, deployment, and current trends: bit.ly/47iexAz v/@Hesamation

10 171 835 49K 905

Download Image

Pinaki Dasgupta ,MBA ✨ @Seanku

3 weeks ago

Parathinker : Native Parallel Thinking is the New Paradigm to Scale LLM Test-time Compute.😇

Rohan Paul @rohanpaul_ai

3 weeks ago

Parathinker : Native Parallel Thinking is the New Paradigm to Scale LLM Test-time Compute.😇

28 165 944 52K 820

Download Image

0 0 0 39 0

Pinaki Dasgupta ,MBA ✨ @Seanku

3 weeks ago

Bootstrapping Task Spaces for Self-Improvement. Exploratory Iteration directly exploits the recurrent structure of self-improvement tasks to train LLMs to perform multi-step self-improvement at inference-time.arxiv.org/abs/2509.04575

0 0 0 69 0

Guilherme Penedo @gui_penedo

3 weeks ago

> we've hit a data wall > pretraining is dead Is it? Today we are releasing 📄 FinePDFs: 3T tokens of new text data for pre-training that until now had been locked away inside PDFs. It is the largest permissively licensed corpus sourced exclusively from PDFs.

27 152 2K 132K 888

Download Image

Pinaki Dasgupta ,MBA ✨ @Seanku

3 weeks ago

Google's AI system uses LLMs & tree search to automate writing expert-level scientific software. invents novel methods, outperforming humans in bioinformatics (40 new methods for single-cell analysis), epidemiology, geospatial analysis. Similar 2 OpenAI !!

Orion (e/acc) @SerendipitousOr

3 weeks ago

1 0 1 4K 2

0 0 0 62 0

Pinaki Dasgupta ,MBA ✨ @Seanku

3 weeks ago

In summary, Sonoma Dusk Alpha & Sonoma Sky Alpha are likely experimental or beta versions of Grok 4 or its smaller variants, deployed under a pseudonym for testing or strategic reasons, with capabilities aligning closely w/ Grok’s advanced reasoning and large context handling.

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭 @elder_plinius

3 weeks ago

71 62 1K 228K 238

Download Image

0 0 0 205 0

Ahmad @TheAhmadOsman

3 weeks ago

there is a lot of MONEY in this add `/.json` at the end of any Reddit link > and get the entire thread > including all replies to the n-th depth > and all the metadata > as JSON and then use LLMs to extract/analyze/etc you can make so much $$$ from niche subreddits

225 515 11K 1.0M 19K

Avi Chawla @_avichawla

4 weeks ago

95% of enterprise AI projects don't reach production. MIT’s 2025 report revealed it. Here's what the remaining 5% do differently:

15 46 540 94K 824

Adam Silverman (Hiring!) 🖇️ @AtomSilverman

4 weeks ago

I summarized everything that happened this week in AI Agents from HappyRobot, AgentOps, YC, OpenAI, Xpander, TextQL, Rork, Copilot Kit, Nautilus, Agno, E2B, Questera, Cursor, Manus, LlamaIndex, & more. 🧵 (save for later)