Surya

Agents for Pentesting your Agents, Apps and networks #redreaming #Agents #Pentesting

[Surya @surya66]

The fastest coding agent ever becomes more faster.

[SpaceX @SpaceX]

SpaceX has exercised the option to acquire @cursor_ai in an all-stock transaction with the goal of building the world’s most useful AI models.

For the past few months, SpaceXAI has been jointly training a model with Cursor, which will be released in Cursor and Grok Build soon.

We look forward to working closely with the Cursor team to advance our frontier AI capabilities

[Surya @surya66]

@SarvamAI @hcltech @BessemerVP @khoslaventures @peakxvpartners Congratulations @SarvamAI

[Surya @surya66]

If you want to save tokens on Fable5 use your Claude code with Headroom https://t.co/s0lJl2TIsK

[Surya @surya66]

AWS keys, JWTs, credit cards, API tokens — flagged automatically. Your agent never knows. Your logs stay clean.

Drop `dscan scan ./` into CI as a gate. Exits 1 on high-severity findings.

Open source. MIT. Python 3.11+

[Surya @surya66]

Introducing dscan — security observability for AI agents 🔍

One decorator. That's all it takes:

@watch
async def my_agent(task: str):
... # unchanged

[Surya @surya66]

✅ Every tool call traced
✅ Secrets auto-redacted before touching disk
✅ Risky prompts & MCP configs statically scanned
✅ Local dashboard to inspect it all

[Surya @surya66]

Can’t wait to get access and try in our https://t.co/oDfGJNsMZT

[OpenAI @OpenAI]

Introducing Daybreak: frontier AI for cyber defenders.

Daybreak brings together the most capable OpenAI models, Codex, and our security partners to accelerate cyber defense and continuously secure software.

A step toward a future where security teams can move at the speed defense demands.

[Garry Tan @garrytan]

This is cool

[Surya @surya66]

Personal AI is next big thing

[pradeep @pradeep24]

tested out @antirez' ds4.c this morning. so impressive and delivers.

on a M3 max, 128GB, stock ds4 settings:
- 14–15 t/s at 62K pre-filled actual coding conversation
- memory usage was flat during gen ~85GB res
- disk cache is ~8GB for a full 100K context window
- thermals were normal, light fan activity
- inference server is rock solid so far

biggest constraint: anytime there's a compact, we pay the wait-time price of a fresh prefill (~1min per 10k context) before we are back in action.

sequential inference + multiple agents in parallel performance is unclear, will report back.

I'm so amped.

[Surya @surya66]

Open source is unstoppable

[Bindu Reddy @bindureddy]

🚨 OPEN SOURCE AI IS LITERALLY UNSTOPPABLE 🚨

The legendary founder of Redis (Antirez) just dropped ds4 - a custom native inference engine built specifically for DeepSeek v4 Flash

This is earth shattering! Here is why:

DeepSeek v4 Flash is a quasi-frontier model with a massive 1M context window

You can now run it LOCALLY on a 128GB Mac using specialized 2-bit quantization

The architecture is reimagined—he moved the KV cache from RAM directly to the SSD disk! 🤯

We already know DeepSeek v4 Flash is insanely good for agentic loops - Now you don't even need the cloud to run it

Closed-source labs are burning tens of billions on massive GPU clusters while single brilliant developers are running frontier-level AI on laptops!

They told us open-source would be worthless against trillion-dollar monopolies

Instead, pure hacker culture + incredible open-weight models are completely rewriting the rules

Open Source will ALWAYS win 💕

[Surya @surya66]

We are already in the Anthropic’s Cyber Verification Program and applied for GPT-5.5-Cyber, we harness the power of multi model so looking forward to getting access. @sama

[Sam Altman @sama]

we're starting rollout of GPT-5.5-Cyber, a frontier cybersecurity model, to critical cyber defenders in the next few days.

we will work with the entire ecosystem and the government to figure out trusted access for cyber; we want to rapidly help secure companies/infrastructure.

[Surya @surya66]

This is first like of defence before teams does regular scans with their security tooling in the pipelines

[Claude @claudeai]

Claude Security is now in public beta for Claude Enterprise customers.

Claude scans your codebase for vulnerabilities, validates each finding to cut false positives, and suggests patches you can review and approve.

[Surya @surya66]

Manual pentesting was the right answer when humans wrote the code.

Now AI writes 46% of it. Code ships in seconds. Pentests still take two weeks and run once a year.

The same AI breaking pentesting is what fixes it

[Surya @surya66]

@aakashgupta Very valuable

[Surya @surya66]

Got into the Anthropic’s Cyber Verification Program 🚀

Big step forward for what we’re building at CyberImmune & DeepScan — AI-native, agentic pentesting for the next generation of software.

More soon. 🔐

#CyberSecurity #AI #StartupJourney

[Surya @surya66]

Must watch podcast

[Dwarkesh Patel @dwarkesh_sp]

Did a very different format with @reinerpope – a blackboard lecture where he walks through how frontier LLMs are trained and served.

It's shocking how much you can deduce about what the labs are doing from a handful of equations, public API prices, and some chalk.

It’s a bit technical, but I encourage you to hang in there - it’s really worth it.

There are less than a handful of people who understand the full stack of AI, from chip design to model architecture, as well as Reiner. It was a real delight to learn from him.

Recommend watching this one on YouTube so you can see the chalkboard.

0:00:00 – How batch size affects token cost and speed
0:31:59 – How MoE models are laid out across GPU racks
0:47:02 – How pipeline parallelism spreads model layers across racks
1:03:27 – Why Ilya said, “As we now know, pipelining is not wise.”
1:18:49 – Because of RL, models may be 100x over-trained beyond Chinchilla-optimal
1:32:52 – Deducing long context memory costs from API pricing
2:03:52 – Convergent evolution between neural nets and cryptography

[Rohit @rohit4verse]

An OpenAI engineer burned 210 billion tokens in one week. That's Wikipedia, 33 times over.

they have investor money to burn. you don't.

the fix takes 2 minutes:
> give this repo to your agent
> tell it to port the patterns over

Save this before you go bankrupt

[N Chandrababu Naidu @ncbn]

తెలుగు భాష గొప్పదనాన్ని వివరిస్తూ.. ఎంతో శ్రావ్యంగా గానం చేసిన ఈ 6వ తరగతి చిన్నారిని మనసారా అభినందిస్తున్నాను. ఈ పాట సాహిత్యాన్ని వింటుంటే ప్రజలకు తెలుగు భాష మీద ఉన్న మమకారం స్పష్టమవుతోంది. అలాంటి తెలుగును పాలకులు కనుమరుగు చేయాలనుకోవడం ప్రజల అభీష్టానికి వ్యతిరేకం.. దారుణం.

[Surya @surya66]

Salute to the GO driver.

[GO Transit @GOtransit]

A quick-thinking GO bus driver is being hailed as a hero after coming to the rescue of a young woman, who was being stalked after dark. We’re so grateful to Bryan for going beyond the call of duty. Here’s what transpired: https://t.co/1MbSGaCiVP