HackerTrans
TopNewTrendsCommentsPastAskShowJobs

henrysg

no profile record

Submissions

Show HN: Shepherd – open-source tools to manage self-hosted software deployments

trustshepherd.com
5 points·by henrysg·w zeszłym roku·0 comments

SOC2 compliance is a waste of time

blog.trustshepherd.com
5 points·by henrysg·2 lata temu·2 comments

So you want to offer self-hosted deployment: how should you do it?

blog.trustshepherd.com
2 points·by henrysg·2 lata temu·0 comments

Should self-hosted deployments be vendor or customer managed?

blog.trustshepherd.com
2 points·by henrysg·2 lata temu·3 comments

Sam Altman-led nuclear startup signs major AI power supply deal

ft.com
1 points·by henrysg·2 lata temu·0 comments

Why is supporting customer-hosted deployments still so hard?

plum-variation-418661.framer.app
6 points·by henrysg·2 lata temu·0 comments

The merits of BYOC vs. SaaS hosting software

jack-vanlightly.com
1 points·by henrysg·2 lata temu·0 comments

Show HN: LinkedIn Hype Crew

16 points·by henrysg·2 lata temu·19 comments

Show HN: Podial – Generate engaging educational podcasts from any documents

7 points·by henrysg·2 lata temu·0 comments

The products enabled by multi-agent LLM frameworks

blog.context.ai
1 points·by henrysg·2 lata temu·0 comments

Retool State of AI

retool.com
1 points·by henrysg·2 lata temu·0 comments

Dramatically bigger model training clusters are coming

semianalysis.com
2 points·by henrysg·2 lata temu·1 comments

Proving ROI for LLM Products

blog.context.ai
1 points·by henrysg·2 lata temu·0 comments

Everyone's talking about the Aschenbrenner AGI interview

blog.context.ai
2 points·by henrysg·2 lata temu·0 comments

Comparing leading LLM multi-agent frameworks

blog.context.ai
3 points·by henrysg·2 lata temu·0 comments

[untitled]

1 points·by henrysg·2 lata temu·0 comments

How to run LLM product evaluations

blog.context.ai
2 points·by henrysg·2 lata temu·0 comments

Ask HN: Tooling to Improve LinkedIn Messaging?

1 points·by henrysg·3 lata temu·3 comments

Show HN: Codebasechat.com – Make a GPT chatbot for any GitHub Repo in 30 seconds

12 points·by henrysg·3 lata temu·2 comments

[untitled]

1 points·by henrysg·3 lata temu·0 comments

comments

henrysg
·w zeszłym roku·discuss
Totally agree
henrysg
·2 lata temu·discuss
I agree, customer-managed self-hosted is definitely more common. I have seen more discussion recently of vendor managed as a middle ground, often called BYOC
henrysg
·2 lata temu·discuss
Nice launch - congrats!
henrysg
·2 lata temu·discuss
You can request account deletion by emailing legal at context dot ai, we have a process for this
henrysg
·2 lata temu·discuss
Totally agree! Its frustrating how hard it is to get profiles from LinkedIn. You can send any resume - a regular PDF, or a LinkedIn download
henrysg
·2 lata temu·discuss
Thank you!
henrysg
·2 lata temu·discuss
Isn't their project called Swarm and yours called Swarms?
henrysg
·2 lata temu·discuss
My takeaways - https://blog.context.ai/is-llm-progress-slowing/
henrysg
·3 lata temu·discuss
> Oddly absent: an over-rehearsed professional release video talking about a revolution in AI.
henrysg
·3 lata temu·discuss
Thanks for the suggestion. I'm not looking for a CRM-like replacement of LinkedIn, I just want a Chrome Extension to make the existing LinkedIn messaging product better
henrysg
·3 lata temu·discuss
Thank you!
henrysg
·3 lata temu·discuss
Fantastic paragraph:

"The point is this: if you accept the premise that regulation locks in incumbents, then it sure is notable that the early AI winners seem the most invested in generating alarm in Washington, D.C. about AI. This despite the fact that their concern is apparently not sufficiently high to, you know, stop their work. No, they are the responsible ones, the ones who care enough to call for regulation; all the better if concerns about imagined harms kneecap inevitable competitors."
henrysg
·3 lata temu·discuss
You're right - the same advice applies to non-AI startups too. I wrote the piece to help the founders who come to me for advice, and they are AI founders in the vast majority of cases.
henrysg
·3 lata temu·discuss
Isn't it more compelling rather than less that my actions back up my words? I previously lived in California!
henrysg
·3 lata temu·discuss
These performance changes seem pretty inevitable when OpenAI is going to continually update the models. Short of versioning every iteration of the model I don't see how developers can avoid these issues. The solution seems to be to implement better telemetry where these APIs are used in production. I've been working on a tool to help with this - www.getcontext.ai
henrysg
·3 lata temu·discuss
Hi HN! I'm Henry, and I recently launched my startup getcontext.ai - its a product analytics tool for AI chat interfaces.

I decided to work on this product after speaking with approximately 100 developers of conversational interfaces, and finding several consistent problems.

This blog outlines the challenges I identified. Would love to hear if they resonate, or if people disagree!