Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
david_shi
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
7 ms
·
91.
▲
by
david_shi
7mo ago
It seems like this is designed for atomic purchases, could it be extended for subscriptions?
92.
▲
by
david_shi
7mo ago
Yeah for sure, I'll add a caveat for public facing skills in production. Mostly interested in gotchas and efficiency improvements learned in shipping those to multiple harnesses, etc.
93.
▲
Ask HN: Have you built a public facing agent skill?
1 points
by
david_shi
7mo ago
|
2 comments
94.
▲
by
david_shi
7mo ago
also: https://en.wikipedia.org/wiki/PRISM
95.
▲
by
david_shi
7mo ago
S -> Z?
96.
▲
China Restricts OpenClaw as Security Fears Grow
(operator.io)
4 points
by
david_shi
7mo ago
|
0 comments
97.
▲
by
david_shi
7mo ago
not knocking this specific implementation in any way, but it's crazy that live OSINT dashboards are now the demo project of choice vs. todo apps
98.
▲
by
david_shi
7mo ago
Where else would you go?
99.
▲
by
david_shi
7mo ago
At some point, the contradiction of "law as something impartial" and "law bends to the whims of power" will need to be resolved.
100.
▲
by
david_shi
7mo ago
Didn't they also effectively cure Hep C? https://www.cnbc.com/2018/04/11/goldman-asks-is-curing-patie...
101.
▲
by
david_shi
7mo ago
If everyone had this attitude we'd still be dying of tuberculosis and countless other diseases.
102.
▲
by
david_shi
7mo ago
Seems insane that profitability so heavily dictates what is researched and what isn't.
103.
▲
by
david_shi
7mo ago
Broadly agree that authorization is the biggest thing to solve for, but it seems like a Django vs. FastAPI distinction. MCP in its default state takes up too much context, creating the new for "MCP as code", which seems a bit circ
104.
▲
How to Build a Personal Intelligence Agency
(operator.io)
1 points
by
david_shi
7mo ago
|
0 comments
105.
▲
by
david_shi
8mo ago
Why don't you just have the agent write scripts against the APIs? The skills-based workflow doesn't confine you to bash only.
106.
▲
by
david_shi
8mo ago
Hyperion is a beautiful book, and prophetic in many ways. RIP to Dan Simmons.
107.
▲
Show HN: OpenClaw Swarm as a Service (YC W20)
(operator.io)
1 points
by
david_shi
8mo ago
|
1 comments
108.
▲
by
david_shi
9mo ago
memetics in action, lots of AI influencers make their living pumping the thing of the week
109.
▲
by
david_shi
9mo ago
> It's like someone is claiming they unlocked ultimate productivity by washing dishes, in parallel with doing laundry, and cleaning their house. In this case you have to take a leap of faith and assume that Claude or Codex will get
110.
▲
by
david_shi
10mo ago
"While LLMs are amazing, they can't run your business by themselves... We ground AI in tight guardrails and deterministic frameworks, optimizing LLMs to deliver enterprise-grade reliability. Trusted. Reliable. Secure." this s
111.
▲
by
david_shi
11mo ago
2 interesting new protocols to check out if you're interested in this topic: 1. https://www.x402.org/ - micropayments for ai agents to access resources without needing to sign up for an api key 2. https://80
112.
▲
by
david_shi
11mo ago
have you built stuff with LLMs before? genuine question because nondeterministic and deterministic workflows are leagues apart in what they can accomplish.
113.
▲
by
david_shi
1y ago
This is awesome, reminds me of the kind of intuition behind PageRank.
114.
▲
by
david_shi
1y ago
rules for thee, not for me
115.
▲
by
david_shi
1y ago
Slack claims it does not train generative AI models on customer data, but this hinges on a narrow definition of “generative.” In practice, Slack does use customer data, including messages, files, and usage behavior, to train global machine
116.
▲
by
david_shi
1y ago
I wonder how much of this data Salesforce can use, a literal goldmine of information
117.
▲
by
david_shi
1y ago
Python monorepo is the biggest surprise in this whole article
118.
▲
Horus: A Protocol for Trustless Delegation Under Uncertainty
(arxiv.org)
3 points
by
david_shi
1y ago
|
0 comments
119.
▲
by
david_shi
1y ago
They have issued a statement but still have not updated the UI. Can't imagine how many users are currently operating as if these messages are deleted.
120.
▲
by
david_shi
1y ago
there's no indication at all on the app that this is happening
More ›