_pdp_

11 ms·

181.
by _pdp_ 4mo ago
The npm cli has bad defaults which you can turn off but they are there I presume for legacy reasons. The secure option is pnpm. The registry is fine. Also on our comment about size differential ... it absolutely can. If I jump from 2 meters
182.
by _pdp_ 4mo ago
Except that the JavaScript / NPM ecosystem is 6-7 times larger than Python and Java / Maven. https://chatgpt.com/share/6a1da751-0d88-832e-ace7-572bc786e0... Check the linked resource which has the actual data
183.
by _pdp_ 4mo ago
Why blame on NPM? Would you blame GitLab if an opensource maintainer was hacked and as a result the repo contains malicious changes? All of these recent incidents is just developers doing stupid things ... like using their compromised devic
184.
by _pdp_ 4mo ago
Oh forgot to mention that each CLI is basically another supply chain issue too. So there is that.
185.
by _pdp_ 4mo ago
At cbk.ai we dynamically load MCPs into the context when the LLM needs them and unload them when finished. The cost for doing this is negligible and it scales well. The good think about MCP is the authentication story. It is almost perfect.
186.
by _pdp_ 5mo ago
A year behind is still very very very good at this price. ;)
187.
by _pdp_ 5mo ago
We will learn the hard way... like always.
188.
by _pdp_ 5mo ago
In the age of abundant intelligence the most valuable thing will be human capital - especially those that give a damn.
189.
by _pdp_ 5mo ago
cbk.ai - nearly 100% eu with a few small exceptions we will finish migrating by the end of Q3
190.
by _pdp_ 5mo ago
Yes I said "To say that LLMs experience emotion is a bit like saying a thermometer feels cold." being sarcastic. The paper spell it out although slightly convolute, i.e. models can exhibit concepts of emotion... and given that the
191.
by _pdp_ 5mo ago
From the paper... > A natural question is whether these emotion concept representations bear any meaningful relationship to human emotional experience. We would urge caution in drawing strong conclusions. > We therefore suggest interp
192.
by _pdp_ 5mo ago
> They are not the cold, calculating robots we were promised. I am not sure how anyone even remotely familiar with how LLMs work can say this. This is a fine-tune job. > The first is our duty to the global poor. I don't think the
193.
by _pdp_ 5mo ago
In 2005 there was one particular cyber security company that comes to mind that had AI claims on their front page. It was a perl script.
194.
by _pdp_ 5mo ago
> I don’t think paying your dues should make you more likely to get through the pipeline. After all, you paid those taxes because you made good money, which is what people come here for. https://www.trumpcard.gov/
195.
by _pdp_ 5mo ago
The docs should start with that with a very clean explanation how it works. Basically first paragraph. :) Otherwise you should expect churn. But also it should really go into some detail how is this different from tool calls with type enfor
196.
by _pdp_ 5mo ago
Maybe I am reading it wrong but I don't think this does what it claim it does or at least how it sounds. Basically this is a tool auto-complete that has a workflow element to it with certain steps that need to happen in certain order.
197.
by _pdp_ 5mo ago
The new Opus feels like a step backwards. More expensive, thinks more, and it does not get the job done.
198.
by _pdp_ 5mo ago
This is way too complex! We have developed a simple API that can produce tokens and events in various formats like jsonl, sse, even csv. Cancelation can happen when the socket is closed when streaming - fully automated - or when you push an
199.
by _pdp_ 5mo ago
Or maybe, just maybe, LLMs do not run deterministicly and that is ok? In the real world almost nothing runs like that - only software and even that has a lot of failures. So perhaps rather than trying to make agents run deterministicly the
200.
by _pdp_ 5mo ago
Well you can communicate organisation structure but not logic or intent. The directory is a tree and the Code is a graph. You can communicate some information by looking at the org chart of a company but it does not really tell you much how
201.
by _pdp_ 5mo ago
VSCode basically any model can edit the 20K file without any issues. The coding harness does not read the entire file at once though. It reads chunks of it so the size does not really matter. What matters is how close are the things the age
202.
by _pdp_ 5mo ago
Coding agents can write ASM. But if you mean writing the actual byte-code that will require a very different approach at a very different level of abstraction that LLMs are not designed to do. Keep in mind that all LLMs are trained first on
203.
by _pdp_ 5mo ago
I am not surprised but this one sticks out... > Models favor monolithic, single-file implementations that diverge sharply from human-written code. Well, all of our code is monolithic with some files close 20K lines of code and we do use
204.
by _pdp_ 5mo ago
The so called AI job loses are not due to AI. I don't think there is anyone out there to argue otherwise. In a year but probably sooner, when software systems start collapsing, and they will, hiring in tech sector will skyrocket. In fa
205.
by _pdp_ 5mo ago
About two years ago I was using the term "agentic engineer" to describe someone who builds AI agents - not a vibe coder. Agentic Engineer does not make much sense to be applied to a developer. It is weird and confusing to call a w
206.
by _pdp_ 5mo ago
Thanks. This is actually interesting. The only downside is that it only works within AWS.
207.
by _pdp_ 5mo ago
Git is already versioned, S3 support versioning and any file copied into the sandbox, is well a copy, so I am not sure what is the angle here. Other than that it looks cool!
208.
by _pdp_ 5mo ago
The reason this blog post does not come with any concrete examples how to use this enablement for useful and constructive things tells you something very important - it is a toy and they do not know who and how they will use it. It is cool
209.
by _pdp_ 5mo ago
K2.6. My experience is anecdotal but GLM was able to complete the task more throughly and quicker.
210.
by _pdp_ 5mo ago
We just migrated an AI agent from Kimi to GLM and frankly I am surprised by the results. It feels premium. However, both Kimi and GLM can end up in doom loops so be careful how you use them. Without a proper harness the agent can easily get

More ›