HackerTrans
TopNewTrendsCommentsPastAskShowJobs

ashed96

no profile record

Submissions

[untitled]

1 points·by ashed96·2 mesi fa·0 comments

Show HN: LLM-Wiki but for Early Founders

github.com
6 points·by ashed96·3 mesi fa·2 comments

Ask HN: Is Python's dynamic typing a liability when generating code with AI?

1 points·by ashed96·8 mesi fa·3 comments

comments

ashed96
·3 mesi fa·discuss
GTM Engine turns Claude Code into your sales team. Eight opinionated slash-command skills that run your entire go-to-market motion — pipeline tracking, daily rituals, call analysis, lead scoring, cold outreach — all from your terminal.
ashed96
·8 mesi fa·discuss
Glad to hear this, finally can use Claude for my projects!

And missing sources:

https://www.anthropic.com/news/claude-in-microsoft-foundry

https://www.anthropic.com/news/microsoft-nvidia-anthropic-an...
ashed96
·8 mesi fa·discuss
Junior makes random mistakes. LLM makes consistent mistakes.

After a few weeks you know the patterns: where it's reliable, where it hallucinates.
ashed96
·8 mesi fa·discuss
You're right - poor wording. Static vs dynamic is what I meant.
ashed96
·8 mesi fa·discuss
Suddenly feeling better about our 99.9% uptime SLA.

When even Cloudflare goes down, nobody can blame the little guys.
ashed96
·8 mesi fa·discuss
No specific structure - you'll need to experiment based on your project.

I treat it like briefing a senior engineer. Example:

"Currently we have X, Y, Z but it has ABC limitation. We need to support FOOBAR. What are our options?"

Agent suggests 3 approaches.

"Go with Approach 1 but remove the Z part. Implement it."

Try being specific about constraints and current state. The agent fills in the implementation details, but you drive the decisions.
ashed96
·8 mesi fa·discuss
From that standpoint, it's definitely happening already. But removal is easier too now. The real question is whether teams will have the discipline to delete. Most won't, but those who do will build better products.
ashed96
·8 mesi fa·discuss
AI made us 10x more productive. Management noticed. Now we are expected to ship 10x more features.

The free time was a lie we told ourselves.
ashed96
·8 mesi fa·discuss
I'm building a startup myself and we vibe code everything, except core/complex parts where AI struggles.

My workflow:

1. Give the agent detailed task + relevant file contexts (so it doesn't waste time searching)

2. Guide it with continuous feedback when it veers off

3. Jump in manually if it's going completely wrong direction

I use Claude Code for the agent work, Cursor for manual edits.

Key insight: AI coding works as well as your architecture does. I make all architecture decisions, file structure, organization -- AI just writes the implementation logic.

Clean architecture = better AI output.

I believe This is becoming the norm because it gives startups massive velocity advantage. That's probably why your CTO is pushing it.

Career impact? As a senior who understands what AI writes, you're gaining superpowers, not losing skills. You were previously bottlenecked by implementation speed and needed other engineers. Now you can be a 10x engineer - as long as you can envision the architecture, AI helps you build it faster.

The key is knowing when to guide vs when to take over. That judgment comes from experience.
ashed96
·8 mesi fa·discuss
Once you've gone through the learning resources others shared, LeetCode has a dedicated concurrency section for hands-on practice: https://leetcode.com/problemset/concurrency/

(I actually authored a few of those problems - they cover the classic scenarios like producer-consumer, dining philosophers, etc.)
ashed96
·8 mesi fa·discuss
It had been up for me, from Asia.
ashed96
·8 mesi fa·discuss
Couldn't relate more.

I built this Chrome extension using similar logic to youtube-dl: https://chromewebstore.google.com/detail/audio-only-youtube/...

Eventually couldn't keep up with maintenance and leave it, sigh.
ashed96
·8 mesi fa·discuss
My WHY: I need money to pay rent. My company's WHY: Generate shareholder value.

There, saved everyone $97. You're welcome.
ashed96
·8 mesi fa·discuss
Nice read though!

Also checkout https://www.youtube.com/watch?v=iT7kq-R3Gjc too. It's absolute gold for GEO/AEO.
ashed96
·8 mesi fa·discuss
What's your take on using AI-generated content for GEO? If it follows E-E-A-T and adds genuine value, does the origin matter for AI visibility?
ashed96
·8 mesi fa·discuss
In theory, maybe to some extent yes - unique usernames could beat reused emails.

But let's be real - nobody actually does that.
ashed96
·8 mesi fa·discuss
All marketers ~~are liars~~ tell stories.
ashed96
·8 mesi fa·discuss
Still on youtube-dl here - didn't know yt-dlp was the actively maintained fork.
ashed96
·8 mesi fa·discuss
IIRC there's pre-processing (embedding/tokenization?) before feeding images to LLMs?

Hit this issue optimizing LLM request times. Ending up lowering image resolution. Lost some accuracy but could bear that.
ashed96
·8 mesi fa·discuss
What metrics are you tracking specifically?