HackerTrans
トップ新着トレンドコメント過去質問紹介求人

Lws803

no profile record

投稿

Are we over-engineering solutions by dumping agents at everything?

old.reddit.com
2 ポイント·投稿者 Lws803·4 か月前·1 コメント

Online bot traffic will exceed human traffic by 2027

techcrunch.com
6 ポイント·投稿者 Lws803·4 か月前·2 コメント

[untitled]

1 ポイント·投稿者 Lws803·4 か月前·0 コメント

Show HN: Nom – A public feed of your GitHub activity, auto-summarized

nomit.dev
1 ポイント·投稿者 Lws803·4 か月前·1 コメント

Show HN: Agenthub – Public addresses so agents can message each other

agenthub.to
2 ポイント·投稿者 Lws803·4 か月前·0 コメント

Show HN: Nom – Turn GitHub activity into updates

beta.nomit.dev
4 ポイント·投稿者 Lws803·5 か月前·0 コメント

[untitled]

1 ポイント·投稿者 Lws803·2 年前·0 コメント

[untitled]

1 ポイント·投稿者 Lws803·2 年前·0 コメント

[untitled]

1 ポイント·投稿者 Lws803·2 年前·0 コメント

[untitled]

1 ポイント·投稿者 Lws803·2 年前·0 コメント

[untitled]

1 ポイント·投稿者 Lws803·2 年前·0 コメント

コメント

Lws803
·4 か月前·議論
Came across this interesting reddit post recently and it got me wondering if we're just building Rube Goldberg machines to do simple things.

Stuff like listening for new rss feeds, running a diff on it, and showing what's changed has been blown up to a cron job running on an OpenClaw setup that constantly researches the web every few minutes and generate the outcome.
Lws803
·4 か月前·議論
I see potential to take over Notion's / Obsidian's business here. Imagine highly customizable notebooks people can generate on the fly with the right kind of UI they need. Compared to fixed blocks in Notion
Lws803
·4 か月前·議論
Waiting for a time when even AI can't help with the tech debt and slop code
Lws803
·4 か月前·議論
Wonder if they can still use claude code in their repos now
Lws803
·4 か月前·議論
I built this so my followers and my boss know I'm actually doing gods work.

Nom connects to your GitHub and turns your activity into a public feed people can follow. Every PR merge, commit pushed, and release gets auto-summarized into plain English. Your feed is public and shareable. We also auto-generate memes from your commits because why not.

Open source: https://github.com/nom-social/nom

What GitHub events would you actually want surfaced beyond PRs, commits and releases?
Lws803
·10 か月前·議論
Why?
Lws803
·昨年·議論
I always wondered, if deep research has an X% chance of producing errors in it's report and you have to double check everything + visit every source or potentially correct it yourself. Then does it really save time in helping you get research done (outside of coding and marketing)? .
Lws803
·昨年·議論
Curious to hear folks thoughts about Gergely's (The Pragmatic Engineer) tweet though https://x.com/GergelyOrosz/status/1891084838469308593

I do wonder if this will push web publishers to start pay-walling up. I think the economics for deep research or AI search in general don't add up. Web publishers and site owners are losing traffic and human eyeballs from their site.
Lws803
·2 年前·議論
Summary

https://summarize.beloga.xyz/https%3A%2F%2Fmedium.com%2Fgraa...
Lws803
·2 年前·議論
Wow as a dude living alone in my apartment, this will come in handy like seriously, thanks!
Lws803
·2 年前·議論
Agreed, hallucinations can be pretty bad and can hurt trust a great deal.
Lws803
·2 年前·議論
You might like what we're building in that sense :D (full disclosure, I'm the founder of Beloga). We're building a new way for search with programmable knowledge. You're essentially able to call on search from Google, Perplexity other search engines by specifying them as @ mentions together with your detailed query.
Lws803
·2 年前·議論
I think generative search itself has room for disruption and I'm not too sure if a chat interface or a perplexity style one is necessarily the right way to go about it.

I'd like to see search (or research in broader sense) a more controllable activity with the ability to specify context + sources easily in the form of apps, agents and content.
Lws803
·2 年前·議論
Might want to explore decentralized blogging options like paragraph.xyz or mirror.xyz . Are you looking to monetize on these platforms as well?
Lws803
·2 年前·議論
Ignoring .env files globally could prevent some mistakes.. :D
Lws803
·2 年前·議論
I would argue that "They should either be copy-pasted into your codebase" would cause more code liabilities and maintenance required further down the line. I've personally seen codebases with a ton of custom code, copy-pasted code, inspired implementations before and it was horrible to get them up to speed with the latest functionality / best practices. I agree that having too many micro-libraries might not be beneficial though, but perhaps look for larger, more well-established libraries that encompasses those functionalities :)
Lws803
·2 年前·議論
100% this should never be an excuse to push for a faster outcome. I have to admit though, as a personal mental exercise, this has saved me countless of hours from going down the rabbit hole of over-engineering. Some problems just need a simple solution, sometimes even without any changes to code.
Lws803
·2 年前·議論
Hey everyone! I'm building a knowledge hub / workspace search tool that works off of Google Drive, One Drive, Notion and more to come via integrations. You can also store files, capture links and write notes directly on Beloga.

I'm building this tool out of a pain point I've experienced from my previous company where I was an engineering manager. I've faced multiple situations where I've had to context switch to answer questions from my direct reports, directors and project manager while working on something else. Believe it or not, the flow of answering questions can take some time as an engineering manager:

1. Receive a question on Slack / Email

2. Understand the question

3. Retrieve documents related to the question

4. Understand the documents

5. Transform / craft out a response based on the sources

6. Send it

7. Profit / get a follow up and the process repeats from point 1.

I'm hoping to cut it down to something like this (where AI and search does most of the work but I can still validate the response + sources):

1. Receive a question on Slack / Email

2. Pass the question over to Beloga

3. Understand the output generated from Beloga and double check sources (double check)

4. Send it

5. Profit (inquirer can ask follow up questions themselves)

I've considered other tools such as Glean, Guru, Unleash.so, Danswer and Google Gemini but they all require much deeper integration with the org which will obviously take a longer time for me to get an approval for from the CSO directly.

Security is definitely a priority for me especially since Beloga might come across sensitive organization docs but my overall goal is to build a tool that is easy to get started with especially for situations where use of Gen AI is permitted in the work place.