Hacker News: rd42

New comment by rd42 in "Tonone-AI/elephant: Persistent memory for Claude Code. Never forget a session"

rd42 — Tue, 14 Apr 2026 13:51:29 +0000

loved the name ! elephants never forget

New comment by rd42 in "I still prefer MCP over skills"

rd42 — Fri, 10 Apr 2026 10:35:26 +0000

I think the key problem is that usage of MCP servers is not 'baked' into the LLM training - but API's and CLI's are already a part of training. So to use your MCP server, the LLM has to use additional intelligence which could have been used to do the actual work instead.

It Is Time

rd42 — Thu, 14 Nov 2024 19:11:59 +0000

Article URL: https://blog.nishantsoni.com/p/it-is-time

Comments URL: https://news.ycombinator.com/item?id=42139976

Points: 1

# Comments: 0

New comment by rd42 in "Nvidia releases NVLM 1.0 72B open weight model"

rd42 — Wed, 02 Oct 2024 08:23:36 +0000

I think the only relevant part to note here is that this model showed improved text-only performance after multimodal training. Wonder if this translates to Llama models also ? Is it possible to extend Llama 3.1 405b with multi-modal training to create another SOTA large model ?

New comment by rd42 in "Show HN: Quilt – Powerful RAG UI for Document QA"

rd42 — Wed, 02 Oct 2024 08:15:06 +0000

Can you explain how is this different from the other offerings. I believe there are quite a few RAG solutions out there, but its not all clear from your description how you guys stand out.

New comment by rd42 in "Ask HN: How to deal with AI generated sloppy code"

rd42 — Tue, 01 Oct 2024 18:35:44 +0000

AI should assist, not replace, skilled coding. It's crucial to blend AI efficiency with human craftsmanship for clean, sustainable code. From my consulting experience, excessive AI reliance results in bloated code that’s a nightmare to maintain. This echoes the problem Java shops faced with class and object overuse, but now it’s amplified.