Mohankumar @Vimokumar

OSCP | Security Researcher | Acknowledged by #Google #Netflix #NokiaSolutionsNetworks #Barracuda @card #Dropmyemail #PuppetLabs #Paymill #Spotify Montreal,Quebec,Canada Joined December 2009

Tweets

660
Followers

304
Following

938
Likes

166

Dan | Machine Learning Engineer @DanKornas

2 weeks ago

If you’re serious about building agentic AI that reasons, retrieves, and acts, this book nails the end-to-end pipeline: LLM foundations → RAG → Knowledge Graphs → productionized agents. It’s practical, code-first, and maps directly to the problems we ship every day. …

42 295 3K 210K 3K

Download Image

Aaron Levie @levie

2 weeks ago

If we have 100X more AI agents than people doing work in a company, it’s actually much more likely that we conform to how agents work best instead of agents conforming to how we work. AI agents fundamentally thrive on context, and the variance in what you get out of them almost…

61 136 596 110K 493

vx-underground @vxunderground

2 weeks ago

Gayfemboy botnet uses the string “meowmeow” as a trigger to activate its backdoor functionality.

30 104 1K 50K 167

Download Image

Aadit Sheth @aaditsh

4 weeks ago

Microsoft literally dropped the best free Generative AI course you’ll ever see

102 494 4K 316K 7K

Download Image

Hamel Husain @HamelHusain

6 months ago

. @sh_reya 's paper confirms what I see in practice 1) Automated evals don't work (without semi-manual human alignment) 2) Most tools don't provide this alignment 3) Automated evals add mostly noise 4) You can only write good evals by looking at data and reacting to failures

19 81 713 76K 669

Download Image

Aaron Levie @levie

a month ago

Great thread. Whatever an AI agent is capable of doing, it can also be tricked into doing. You should assume if an agent can access data, that a user can eventually get that data too. Agent security, access controls, and deterministic guard rails will be critical.

Michael Bargury @mbrg0

a month ago

107 903 9K 1.2M 7K

Download Image

17 72 476 93K 353

Y Combinator @ycombinator

2 months ago

Andrew Ng (@AndrewYNg) on how startups can build faster with AI. At AI Startup School in San Francisco. 00:31 - The Importance of Speed in Startups 01:13 - Opportunities in the AI Stack 02:06 - The Rise of Agent AI 04:52 - Concrete Ideas for Faster Execution 08:56 - Rapid…

41 333 1K 832K 2K

Download Video

Haider. @slow_developer

3 months ago

Andrej Karpathy says self-driving felt imminent back in 2013 -- but 12 years later, full autonomy still isn’t here "there’s still a lot of human in the loop" he warns against hype: 2025 is not the year of agents; this is the decade of agents

59 96 1K 271K 464

Download Video

Steve Derico @stevederico

3 months ago

8 quotes from the @elonmusk interview w/ @garrytan at @ycombinator ai startup school part 1 "you'll see ultimately an economy that is not 10 times more than the current economy...an economy that is thousands of times maybe millions of times bigger than economy today" 🧵 1/8

1 1 2 428 0

Download Video

Aaron Levie @levie

4 months ago

We’re going to look back in a few years and be shocked by how much time it used to take to just do basic business tasks before AI. This is a huge asset for the next generation workforce, because they can just work normally and have no idea why something should be slow.

Jason ✨👾SaaStr.Ai✨ Lemkin @jasonlk

4 months ago

19 10 82 98K 68

Download Video

27 31 211 80K 98

Shruti @heyshrutimishra

4 months ago

🚨 BREAKING: NVIDIA JUST announced roadmap for physical AI, robotics and national-scale AI factories. Here’s a breakdown of the top important announcements: 🧵👇 1. DeepSeek R1 is now 4x faster, setting the standard for AI in inference and reasoning.

102 933 7K 1.7M 4K

Download Video

JS0N Haddix @Jhaddix

4 months ago

== a websec thread == Inspired by @0xTib3rius I wanted to post my taxonomy of the different types of web scanning as i think it's important for people getting into web security to know. I'll frame some of these in their context as it pertains to @PortSwigger 's Burp Suite and…

3 83 466 37K 556

Download Image

Aaron Levie @levie

5 months ago

In recent conversation with IT leaders across a range of industries, the topic of business model transformation has come up more than ever due to AI and AI Agents. Most companies are rapidly thinking through what the impact of their business will be in an AI-First world. Not…

25 37 215 73K 202

Aaron Levie @levie

6 months ago

Did something change in robotics like exactly one month ago to make all of this possible. This is crazy.

637 349 5K 910K 946

Download Video

Sam Altman @sama

7 months ago

no thank you but we will buy twitter for $9.74 billion if you want

12K 18K 228K 33.7M 8K

Aira Security @Aira_Security

11 months ago

Always scan inputs before sending them to an LLM. It helps prevent harmful content, ensures security, and improves accuracy. Stay safe and compliant. #AI #LLM #Security

0 1 1 40 0

Aaron Levie @levie

7 months ago

AI Agents will dramatically expand the size of the software market. Here's how that will work. Traditionally, software companies are stuck within the constraints of existing IT budgets, with IT expenses running in most companies somewhere between 3-7% of revenue (with tech and…

234 130 762 151K 1K

Elon Musk @elonmusk

7 months ago

Oh well

Mario Nawfal @MarioNawfal

7 months ago

Oh well

7K 2K 16K 45.4M 829

Download Image

30K 19K 229K 43.5M 5K

Aira Security @Aira_Security

9 months ago

🚨 𝗟𝗟𝗠-𝗼𝗻𝗹𝘆 𝘀𝗲𝗰𝘂𝗿𝗶𝘁𝘆? 𝗡𝗼𝘁 𝗲𝗻𝗼𝘂𝗴𝗵. 💡 Meet Aira GM3: The multi-modal security product! 🛡️ 🔍 Detecting issues across text, images, audio, and video – because real-world threats don’t stick to one format. #AISecurity #LLMs #GenAI #AiraGM3 #GoForAiraSecurity

0 1 1 27 0

Download Image

Aira Security @Aira_Security

8 months ago

Logits-based attacks: The invisible AI threat As AI becomes central to your business, understanding these vulnerabilities is critical to protecting your models. Learn more: linkedin.com/pulse/logits-b… #AISecurity #LLMThreats #AdversarialAttacks #AiraSecurity