robotswantdata

1,075 karma · joined October 15, 2024

submissions comments

robotswantdata··on Matchlock – Secures AI agent workloads with a Linux-based sandbox

Sandbox won’t be enough, distroless + “data firewall” + audit

robotswantdata··on A sane but bull case on Clawdbot / OpenClaw

Still end up marked. Don’t do it

robotswantdata··on Anthropic AI tool sparks selloff from software to broader market

Huge legal tech business units

robotswantdata··on Moltbook is the most interesting place on the internet right now

But to be useful it’s not in a contained environment, it’s connected to your systems and data with real potential for loss or damage to others.

Best case it hurts your wallet, worse case you’ll be facing legal repercussions if it damages anyone else’s systems or data.

robotswantdata··on Moltbook is the most interesting place on the internet right now

Simon, this is going to produce some nice case studies of your lethal trifecta in action!

robotswantdata··on Amazon cuts 16k jobs

No offence, but this is 5% of what a middle manager does. We automated this task years back without LLMs too.

robotswantdata··on Ask HN: How are you doing RAG locally?

You don’t need a vector database or graph, it really depends on your existing infrastructure , file types and needs.

The newer “agent” search approach can just query a file system or api. It’s slightly slower but easier to setup and maintain as no extra infrastructure.

robotswantdata··on GLM-4.7: Advancing the Coding Capability

I signed up and got access within a few days. They even gave me free credits for a while

robotswantdata··on GLM-4.7: Advancing the Coding Capability

Yes they conveniently forget about disclosing prompt processing time. There is an affordable answer to this, will be open sourcing the design and sw soon.

robotswantdata··on AI will kill all the lawyers

Very much doubt, what you’ll see is it killing off paralegal work.

In most jurisdictions legal advice is a regulated and restricted activity. Qualified lawyers today get themselves into trouble without AI advising on areas they have no right to practice in.

robotswantdata··on How does a "you interview for US company, we do the work" scam work?

You’ll pass the sniff test, they don’t.

Be careful they might try and use your identity to then commit more fraud.

robotswantdata··on GPT-5.2

For a few weeks the Codex model has been cursed. Recommend sticking with 5.1 high , 5.2 feels good too but early days

robotswantdata··on Are consumers just tech debt to Microsoft?

For the average office task they don’t seem far off being competent, at least to the average workers quality.

Ai builder with gpt5 + workflow triggers is very capable already. 1-2 more model generation hops needed plus a bit more “agent” plumbing before its game over for the excel and word jobs.

robotswantdata··on Launch HN: Mosaic (YC W25) – Agentic Video Editing

Gemini 3 would rip through that problem, but equally you could slice the video with existing open source tooling such as FFMPEG then combine with blender for the video curation. Gemini 3 could probably write you the workflow as well.

robotswantdata··on Tesla Robotaxi had 3 more crashes, now 7 total

How many crashes per human Uber mile?

robotswantdata··on Building more with GPT-5.1-Codex-Max

Sorry don’t like the max model, feels like it needs a lot more guiding. The plans it writes however are better, so I tried feeding it back in (meta prompt style) and working okay so far. Very large repository.

robotswantdata··on Microsoft's lack of quality control is out of control

Ship loading screen simulator

robotswantdata··on DeepSeek OCR

VLLMs suck at complex layouts and there is a high risk of hallucination. Never use alone for contracts or health data.

robotswantdata··on Andrej Karpathy – It will take a decade to work through the issues with agents

The meme in 22/23 was OpenAI was really “Available AI”

robotswantdata··on Ask HN: Abandoned/dead projects you think died before their time and why?

soon will be able to buy a gigabyte AI Top CXL R5X4. PCI expansion card with up to 512gb RAM over four DIMMs.

robotswantdata··on AMD signs AI chip-supply deal with OpenAI, gives it option to take a 10% stake

Infinite money glitch

robotswantdata··on N8n added native persistent storage with DataTables

Node red is better (imo) but harder for non technical users to use

robotswantdata··on Claude Sonnet 4.5

On pro tier have not been able to trigger the usage cap.

Pro

Local tasks: Average users can send 300-1,500 messages every 5 hours with a weekly limit. Cloud tasks: Generous limits for a limited time. Best for: Developers looking to power their full workday across multiple projects.

robotswantdata··on Claude Sonnet 4.5

Never hit pro quota yet, huge repo. Have multiple projects on the go locally and in cloud.

Feel like this is going to be thr $1000 plan soon

robotswantdata··on Claude Code 2.0

Try giving codex IDE a go, now included with ChatGPT. Had equal frustrations with Claude making bad decisions, in contrast gpt5 codex high is extremely good!

robotswantdata··on Claude Sonnet 4.5

I had similar experience, not good enough yet to come back for the Claude max plan.

Sticking with ChatGPT pro sub and gpt5 codex on high.

robotswantdata··on Instant Checkout and the Agentic Commerce Protocol

https://developers.openai.com/commerce

Yet another protocol

robotswantdata··on 25L Portable NV-linked Dual 3090 LLM Rig

“Easy” to mod to 48gb

robotswantdata··on 25L Portable NV-linked Dual 3090 LLM Rig

Heating

robotswantdata··on Introducing Stargate UK

Unlikely. Don’t believe the online discourse to be representative of the people.

See: https://www.telegraph.co.uk/politics/2025/06/25/scottish-ind...

← PreviousPage 5 of 6Next →