Hacker News: garkotipankaj

New comment by garkotipankaj in "Agent Skill to Force Docs in ASD-STE100 Simplified Technical English"

garkotipankaj — Thu, 06 Aug 2026 00:40:53 +0000

AH. https://github.com/computer-reinvention/trie

Sorry. A missing hyphen!

New comment by garkotipankaj in "If AI Outputs Aren't Speech, Who Has to Prove They're Human?"

garkotipankaj — Mon, 03 Aug 2026 20:45:53 +0000

btw "non-deterministic text generator" also describes humans perfectly. if it didn't you'd be less stupid.

New comment by garkotipankaj in "If AI Outputs Aren't Speech, Who Has to Prove They're Human?"

garkotipankaj — Mon, 03 Aug 2026 20:44:03 +0000

if someone asks their assistant to send you an email on their behalf, whose speech was it?

New comment by garkotipankaj in "Agent Skill to Force Docs in ASD-STE100 Simplified Technical English"

garkotipankaj — Fri, 31 Jul 2026 06:01:17 +0000

Agent skills work... sometimes - depends on the agent's mood really.

I tried to solve this in a rather opposite way.

https://github.com/computerreinvention/trie

The documentation is updated and recording the intent behind all changes is enforced by a commit-gate. At the symbol level.

New comment by garkotipankaj in "Knoppix"

garkotipankaj — Wed, 01 Jul 2026 01:14:58 +0000

Same story! I discovered Porteus (based on slackware) and used it from a 8gb sandisk usb for years. It shipped with xfce and was a blast to tinker with. I learned to compile software from source for it because the package management was non-existent. Taught me a lot. AAAND it's still live: https://www.porteus.org

New comment by garkotipankaj in "Show HN: Infera – agentic CLI for inferring and provisioning cloud infra"

garkotipankaj — Thu, 22 Jan 2026 12:30:43 +0000

Hi there! I recently had to learn terraform to setup the infra for a new place I joined. Claude Code helped A LOT - but there were instances where it was clear that had I not been experienced enough, I would be banging my head against the wall for days. This is where I got the idea.

The tool itself is a thin wrapper over terraform and uses the Claude Agent SDK combined with a 80+ best practice architectural templates to determine the best way to get your project on your preferred cloud provider. Would love to know if this interests anyone else.

The interface is taken direclty from terraform and has the exact same subcommands - init, plan and apply. You can also run `infera deploy` to do all three at once.

Show HN: Infera – agentic CLI for inferring and provisioning cloud infra

garkotipankaj — Thu, 22 Jan 2026 12:24:57 +0000

Article URL: https://github.com/computer-reinvention/infera

Comments URL: https://news.ycombinator.com/item?id=46718346

Points: 2

# Comments: 2

New comment by garkotipankaj in "Show HN: I built an AI SuperConnector for founders and VCs"

garkotipankaj — Wed, 05 Nov 2025 15:15:12 +0000

On first glance, this seems very gimmicky. The whole wolf fursona thing does not make sense. What's the logic here? That wolves are social animals? I digress.

There is nothing on the website that works to remove this initial skepticism. Your AI maybe actually wonderful but atleast for me, it is hard to trust it - it is purely because you have chosen to present it in a way that is basically a cosplay. If I trust an AI to connect me with VCs and potential employers / clients, I definitely expect it to be a lot more serious in presentation. Right now it looks like a furry's personal website. No shade to furries but it simply does not win my trust. Please re-consider the way you are branding this if you want people to use it like a serious tool.

Having said that, the core idea seems interesting to me. LinkedIn has become a cesspool of AI slop. People writing posts with AI and then people commenting on those posts with AI. Funny enough, AI itself seems to be the thing that will end up saving us from that non-sense.

I was going to try this but I do not have WhatsApp. Most people in North America don't. So I can't even access your product. Thanks for wasting my time.

New comment by garkotipankaj in "Show HN: ChainFactory – Run Structured LLM Inference with Easy Parallelism"

garkotipankaj — Tue, 30 Jul 2024 02:12:47 +0000

Ah. Just read the rules. Version updates are not ok unless it's a major overhaul.

Will this be removed? I still desperately need feedback :(

Show HN: ChainFactory – Run Structured LLM Inference with Easy Parallelism

garkotipankaj — Tue, 30 Jul 2024 01:48:07 +0000

Hi HN!

Disclaimer: I submitted another post about ChainFactory a few days ago.

Here's what has changed since:

- Added hash based caching of auto-generated prompts and masks.

- Did some internal restructuring and cleanup.

- Updated the order in which README doc introduces concepts and terminology.

Posting this again because honestly, I am kinda puzzled about what to add/fix/change due to having 0 users and no genuine feedback.

By genuine feedback, I mean feedback from strangers who do not have a social pressure to be polite and pull punches.

Please take a look if you find this interesting and leave a comment. If you think it's an deranged or stupid idea not worth your time, please at least leave a 'no' - I'd still be delighted as it's an honest opinion.

Thanks a lot!

PS: Is it okay to post updates and changes at regular intervals?

Comments URL: https://news.ycombinator.com/item?id=41105402

Points: 2

# Comments: 1

Show HN: ChainFactory – Run Structured LLM Inference with Easy Parallelism

garkotipankaj — Thu, 25 Jul 2024 23:05:35 +0000

hi everyone.

how does moving llm call prompts and output structure definitions away from code into configuration land sound?

would you use something like this if it was stable and well documented enough?

please don't hold back the criticism. i appreciate all feedback (constructive & otherwise).

Comments URL: https://news.ycombinator.com/item?id=41074380

Points: 8

# Comments: 0

New comment by garkotipankaj in "Show HN: Mavy – Personal AI Executive Assistant GPT"

garkotipankaj — Mon, 08 Jan 2024 14:30:44 +0000

I don't see how. On the contrary, we are very much looking forward to the GPT store launch.