lunarcave

10 ms·

91.
by lunarcave 2y ago
How much of this is hype we think? Serious question: are people actually making use of "agents" in their companies to do the tedious work that most vertical agents solve for? I find coding agents (codium/cursor/copilot)
92.
by lunarcave 2y ago
Congrats on the launch. I actually really like how the agency level x capability is displayed on the page. Too many text in / text out abstractions being mislabeled as “agents” these days.
93.
by lunarcave 2y ago
Nice. How does the popularity column get computed?
94.
by lunarcave 2y ago
Perhaps the story that doesn't get told more often is how LLMs are changing how humans operate en masse. When ChatGPT came out, I was increasingly outsourcing my thinking to LMS. It took me a few months to figure out that that's a
95.
by lunarcave 2y ago
Sorry this happened, and your instinct to check LLM responses are correct. With that being said, if you ask me the same question and tell me your app can tolerate downtime, need provisioned compute (EC2) - I'd probably say the same thi
96.
by lunarcave 2y ago
Very nice. Am I correct in assuming that this is true for domain specific (math) problems?
97.
by lunarcave 2y ago
Anthropic has a good definition of this one [1] if you want a more comprehensive view. I've been building something along the same lines [2]. I'd define an agent as a piece of software that can autonomously reason based on the con
98.
by lunarcave 2y ago
If you've played around with LLMs in any serious capacity, you'd figure out that they are pretty good at doing the most sensible thing. So they're pretty good in applications where you have to produce the most sensible thing.
99.
by lunarcave 2y ago
A happy Langfuse customer here! We've been building an agent platform and some of our customers wanted someway to exfil OTEL traces to their own setup. Initially we tried building our own but then realised Languse does exactly what we
100.
Noteshare.space: Securely share your Obsidian notes with one click(noteshare.space)
1 points by lunarcave 2y ago|0 comments
101.
Hackerrank was broken – but now it's harmful(segfaulte.mataroa.blog)
59 points by lunarcave 2y ago|67 comments
102.
by lunarcave 2y ago
> manage the workflow Hey, if you're specifically looking for providing deterministic guardrails around agent calls, I'm solving that particular problem. We're sort of an "RPC layer for tools with reasoning built in&q
103.
by lunarcave 2y ago
The MCP[1] that was announced by Anthropic has a solution to this problem, and it's pretty good at handling this use case. I've also been working on a solution to this problem via long-polling tools. [1] https://github.
104.
by lunarcave 2y ago
Congrats on the launch! Big fan of what you guys are doing.
105.
by lunarcave 2y ago
I think this is a the value prop for AWS Bedrock (no affiliation) AFAIK You get a managed instance where the model data doesn’t get sent to the model provider, but you pay PAYG rates.
106.
by lunarcave 2y ago
People used to fork the JavaScript process according to the number of cores present, but it’s less popular now because most infrastructure is provisioned by vCPUs. So people just provisioned for vCPU=1. Node.js (v8) already offloads io task
107.
by lunarcave 2y ago
If you load something into the LLM context, there's a non-zero chance that it'll be referenced. How are you chunking things? Can you chunk it in a way that sidesteps the problem? It's kind of hard to give generic advice witho
108.
Doing on-device RAG with SQLite and Ollama(inferable.ai)
6 points by lunarcave 2y ago|0 comments
109.
It's a torturous chaos until it isn't(longform.asmartbear.com)
3 points by lunarcave 2y ago|0 comments
110.
by lunarcave 2y ago
This is very cool. Congrats on the launch. Everybody has their own DSL to do the same thing.
111.
by lunarcave 2y ago
Ok, this is cool. I use a few tools that does the same, but with CSVs. I often try to use "AI copilots for mind mapping", but this is good. Rather than trying to pin down the first potential use case, why don't you showcase a
112.
by lunarcave 2y ago
Congrats on the launch. Can I bring an existing DB in and just use this to query the data? I guess I can supply a connstr with the read access?
113.
by lunarcave 2y ago
Ok, this looks cool. I did a small search which yielded some pretty good results. Some feedback: - Longer videos don't tend to fare well. For example, if the video is on the specific topic, then the clipping basically takes in almost t
114.
by lunarcave 2y ago
Congrats on launching! Looks pretty neat. I like the fact that BaseStorage is abstract. `sqlite-vec` had been pretty good at what it does, in case you're looking for a good built-in vector search.
115.
by lunarcave 2y ago
There are a few tools here like e2b and modal. No affiliation. [1] https://e2b.dev [2] https://modal.com I’m in the same space, but we run the code that devs already write using LLMs to orchestrate them.
116.
by lunarcave 2y ago
Also the conflict in Ukraine led to a popularisation of low cost drone mods and their capabilities. Especially in the context of recon and military deployments. So they’ve captured more of a mindshare I guess.
117.
by lunarcave 2y ago
A happy HyperDX customer here. Can't recommend it enough. We wanted something good for tracing and logs, without the price tag we were used to from datadog. We've been pleasantly surprised by how easy it was to set up and start pu
118.
by lunarcave 2y ago
I’m not sure whether this will help, but I’ll tell you what works for me. Zoom out and do things that let you zoom out. There are certain things that you can do in order to get out of the day of your life and realise that there is a whole w
119.
by lunarcave 2y ago
Although I disagree with the arbitrary limit of “one day per feature”, I agree with the broader point that useful constrains drive creativity.
120.
by lunarcave 3y ago
> If your operation is already idempotent in its implementation No, this is not at all what I meant. I don't think there's any value in introducing at-most once semantics to an already idempotent function. I'm definitely n

More ›