What about the kids they intentionally get driven to suicide by keeping the blue bubbles for no other reason than child indoctrination due to bullying from other kids.
I know there can be CTOs who write a lot of code and be good, but I hope to God you're not my CTO because he is a nightmare in this regard right now. We aren't even a startup anymore we have 80 engineers and he pushed 200k lines this week
At this point if someone calls it auto complete they can be written off as a Luddite with nothing valuable to say. The irony being they themselves are being a stochastic parrot by parroting the jargon other people say about llms.
I genuinely think it's part of a psyop. If we bloat all codebases and eventually start printing the models on chips to reduce inference costs by 50-100x they'll take in massive profits from 5M line codebases instead of 350k
Proof they don't nerf it only after testing that the benchmarks there stay the same? So overall performance degrades but they isolate those benchmarks?
Ok so in a situation like regular orchestration you would essentially layout all possible steps the LLM can take in your code in a big orchestration layer, and if it hits the sensitive endpoint the orchestration that can occur past that will block off web search. In the design that is. But for something like a manus style agent where you're outsourcing all the work but allowing it to hit your MCP it just becomes a regular API the LLM can call
I feel like I don't fully understand mcp. I've done research on it but I definitely couldn't explain it. I get lost on the fact that to my knowledge it's a server with API endpoints that are well defined into a json schema then sent the to LLM and the LLM parses that and decides which endpoints to hit (I'm aware some llms use smart calling now so they load the tool name and description but nothing else until it's called). How exactly are you doing the process of stopping the LLM from using web search after it hits a certain endpoint in your MCP server? Or is this referring strictly to when you own the whole workflow where you can then deny websearch capabilities on the next LLM step?
Are there any good docs youve liked to learn about it, or good open source projects you used to get familiar? I would like to learn more
i was at the cvs right next to the extra storage when the helicopters showed up and all the police it was kinda nuts to be so close to an event like this.
i have not written about it due to me being a silent partner in the company and dont want my name publicly attatched to it, but the code and prompts is all more like talking to a buddy is how i use it, i ask it to build specific things then i look through and make changes. For instance a few examples i can give is there is a lot of graph traversal in my data i built, I'm not an expert on graph traversal, so I researched what would be a good algo for my type of data, and then utilized claude to implement the papers algorithm into my code and data structures. I dont have the llm in any steps that the customer interact with (there is some fuzzy stuff but nothing consistently run) but i would say an llm has touched over 90% of the code i wrote. its just an upgraded rubber ducky to me.
If i wasn't experienced in computer science this would all fall apart however i do have to fix almost all the code, but spending 10 mins fixing something is better than 3 days figuring it out in the first place (again this might be more unique to my coding and learning style)
i mean i can state that i built a company wihtin the last year where id say 95% of my code involved using an LLM. I am an experienced dev so yes it makes mistakes and it requires my expertise to be sure the code works and to fix subtle bugs; however, i built this company me and 2 others in about 7 months for what wouldve easily taken me 3 years without the aid of LLMs. Is that an indictment of my ability? maybe, but we are doing quite well for ourselves at 3M arr already on only 200k expense.