News for hackers. Stuff that matters.

kol3x

I initially open-sourced it almost 3 months ago, and this version is more of a finished product. My focus has been on making is easier to pick up, hence the demo that is linked and proper frontend, polished mobile version and a lot of ui/ux and documentation tweaks. The app is my daily companion, and I'd love more people to try it out and share their opinion.

Also, here is a blog post about me building this version: https://kol3x.com/blog/design-before-building-or-my-first-fi...

Show HN: Pawmc a.k.a. talking diary, an LLM chat with compartmentalised memory

https://pawmc-demo.kolaegipet.workers.dev/?demo=1

What pawmc is (from the blog post):It’s an LLM chat app with minimal system prompt and no tools/web-search by design. It stores most important context across your conversations in isolated buckets, called “diaries” and “topics”. This means that when you start a conversation in { diary: personal, topic: books }, the AI receives your first message along with summaries of that diary and that...

( Read More... | 1 comment | on HN )

Show HN: Pawmc a.k.a. talking diary, an LLM chat with compartmentalised memory

https://pawmc-demo.kolaegipet.workers.dev/?demo=1

What pawmc is (from the blog post):It’s an LLM chat app with minimal system prompt and no tools/web-search by design. It stores most important context across your conversations in isolated buckets, called “diaries” and “topics”. This means that when you start a conversation in { diary: personal, topic: books }, the AI receives your first message along with summaries of that diary and that...

( Read More... | no comments | on HN )
Re:Dear LLM, or how I stopped getting generic advice on Thursday October 08, 2026 @07:35AM (#50004589)
I plan a bunch of improvements in the coming weeks, and to post the project as ShowHN again with a reworked v2. Just finished buliding a demo mode, so anyone can get a feel of it without deploying, a link is in the repo's 'about' section.

Anyway, let me know your thoughts.

Re:Ask HN: Who wants to be hired? (September 2026) on Tuesday September 01, 2026 @11:01AM (#49522899)
Location: Tbilisi, Georgia

Remote: Yes

Willing to relocate: Yes

Technologies: JavaScript, TypeScript, Cloudflare Workers, LLM integration, React.

Resume/CV: kol3x.com

Е-mаil: hn-september at kol3x dot com

I started as a fullstack JS engineer and grew into product development at my last company. I owned the ingestion layer - collecting 100k relevant messages/month from 4 platforms through a dirt-cheap serverless back-end on Cloudflare Workers.

We worked with LLMs deeply, especially in code generation research. I was part of a small team developing SpecGEM, a fine-tuned model that generates Cloudflare Workers code from JSDoc comments and is designed to outperform mainstream LLMs at this narrow task. https://dn.institute/research/ai/specgem/

Last month I released a self-hosted LLM assistant on Cloudflare Workers that predictably compounds context across conversations. https://github.com/kol3x/pawmc

Looking for a remote role where I can own work end-to-end. Stack-wise I'm strongest on Cloudflare Workers, JavaScript/TypeScript, practical LLM integration, React.

CV, LinkedIn, and GitHub are also linked at https://kol3x.com

Show HN: Simple self-hosted LLM assistant with user-steered compounding context

https://github.com/kol3x/pawmc

I built a personal LLM assistant on Cloudflare Workers + Durable Objects. You specify a category and topic when starting a new conversation, so the backend maintains a summary for each category/topic - building up as more conversations happen under the same one.There is no complicated RAG, embeddings, or agentic magic, but the category/topic summaries system just works, and I've genuinely found...

( Read More... | no comments | 2 points | on HN )
Re:Ask HN: Who wants to be hired? (August 2026) on Monday August 03, 2026 @11:00AM (#49156687)
Location: Egypt (Russian passport)

Remote: Yes

Willing to relocate: Yes

Technologies: JavaScript, Cloudflare Workers, LLM integration, JSDoc

Résumé/CV: kol3x.com

Email: [email protected]

I started as a fullstack JS engineer and grew into product development at my last company. I owned the ingestion layer - collecting 100k relevant messages/month from 4 platforms through a dirt-cheap serverless back-end on Cloudflare Workers.

We worked with LLMs deeply, especially in code generation. I was part of a small team developing SpecGEM, a fine-tuned model that generates Cloudflare Workers code from JSDoc comments and is designed to outperform mainstream LLMs at this narrow task.

I just released an open-source self-hosted LLM assistant on Cloudflare Workers that predictably compounds context across conversations. https://github.com/kol3x/pawmc

Looking for a remote role where I can own work end-to-end. Stack-wise I'm strongest on Cloudflare Workers, JavaScript, and practical LLM integration.

CV, LinkedIn, and GitHub are also linked at https://kol3x.com

Re:Ask HN: Who wants to be hired? (July 2026) on Friday July 03, 2026 @10:04AM (#48775216)
Location: Earth

Remote: Yes

Willing to relocate: Yes

Technologies: JS, Cloudflare Workers, Product Development, practical LLM integration

Resume & Email: all links can be found at kol3x.com

You already now some companies will optimize for this score but in the opposite way...
Would be nice to filter out "proxied" models in the Workers AI page.
Re:Ask HN: Who is hiring? (November 2025) on Monday November 03, 2025 @11:01AM (#45800486)
DN Institute | Software Engineer | Full Remote (World)

Distributed Networks Institute’s mission is to achieve sustainable technological growth and resilience by leveraging scientific, engineering, and organizational principles of distributed networks. We are a part of a 501(c)3 non-profit incubator in Washington DC called BlockShop.

We just released a paid Software Development challenge with a monetary bounty and a chance to become part of our team. It involves improving a Telegram Connector, deployed on Cloudflare Workers (JS + Python). You can find the full description and more information about challenge program in the Github issue here: https://github.com/1712n/dn-institute/issues/491