Hacker News: tifa2up

New comment by tifa2up in "Show HN: vLLM Studio – A macOS app for using vLLM models"

tifa2up — Sat, 21 Mar 2026 20:07:46 +0000

Interesting project. Curious why Electrobun over Tauri here? Tauri has a much larger ecosystem and rust based for improved performance.

Gemini 2 Is the Top Model for Embeddings

tifa2up — Wed, 11 Mar 2026 09:41:53 +0000

Article URL: https://agentset.ai/blog/gemini-2-embedding

Comments URL: https://news.ycombinator.com/item?id=47333508

Points: 5

# Comments: 0

The Next Coding Interface Is a Canvas

tifa2up — Fri, 06 Mar 2026 14:28:07 +0000

Article URL: https://blog.abdellatif.io/the-next-coding-interface-is-a-canvas

Comments URL: https://news.ycombinator.com/item?id=47275296

Points: 2

# Comments: 0

Hackable Software

tifa2up — Wed, 11 Feb 2026 18:47:33 +0000

Article URL: https://blog.abdellatif.io/hackable-software

Comments URL: https://news.ycombinator.com/item?id=46979039

Points: 1

# Comments: 0

New comment by tifa2up in "Ask HN: What are you working on? (February 2026)"

tifa2up — Mon, 09 Feb 2026 12:11:52 +0000

https://agentset.ai/

Open-source RAG infrastructure.Every team I talk to has the same experience: RAG works in the demo, breaks in production.

We handle ingestion through retrieval with optimizations baked in. 97.9% on HotpotQA vs 88.8% for standard RAG. Model-agnostic, 22+ file types, built-in citations, MCP server. MIT licensed.

https://github.com/agentset-ai/agentset

Do things like Oh My OpenCode work?

tifa2up — Wed, 04 Feb 2026 15:42:56 +0000

Article URL: https://github.com/code-yeongyu/oh-my-opencode

Comments URL: https://news.ycombinator.com/item?id=46887198

Points: 1

# Comments: 0

Show HN: EasyClaw – Seamless Installation for OpenClaw

tifa2up — Mon, 02 Feb 2026 09:34:18 +0000

Article URL: https://easyclaw.com

Comments URL: https://news.ycombinator.com/item?id=46854078

Points: 3

# Comments: 3

New comment by tifa2up in "Ask HN: Share your personal website"

tifa2up — Thu, 15 Jan 2026 09:45:52 +0000

https://abdellatif.io

Semantic Thesaurus

tifa2up — Fri, 19 Dec 2025 15:00:46 +0000

Article URL: https://thesaurus.abdellatif.io/

Comments URL: https://news.ycombinator.com/item?id=46326564

Points: 2

# Comments: 0

Vector DB Comparison

tifa2up — Mon, 08 Dec 2025 09:04:26 +0000

Article URL: https://agentset.ai/blog/best-vector-db-for-rag

Comments URL: https://news.ycombinator.com/item?id=46190040

Points: 1

# Comments: 0

Always Do a Work Trial

tifa2up — Mon, 01 Dec 2025 16:44:43 +0000

Article URL: https://blog.abdellatif.io/always-do-a-work-trial

Comments URL: https://news.ycombinator.com/item?id=46109577

Points: 2

# Comments: 0

New comment by tifa2up in "28M Hacker News comments as vector embedding search dataset"

tifa2up — Sat, 29 Nov 2025 09:01:41 +0000

https://agentset.ai/leaderboard/embeddings good rundown of other open-source embedding models

Opus 4.5 is the best model for RAG

tifa2up — Tue, 25 Nov 2025 13:51:06 +0000

Article URL: https://agentset.ai/blog/opus-4.5-eval

Comments URL: https://news.ycombinator.com/item?id=46045760

Points: 2

# Comments: 0

Gemini 3 vs. GPT 5.1 for RAG

tifa2up — Wed, 19 Nov 2025 17:10:11 +0000

Article URL: https://agentset.ai/blog/gemini-3-vs-gpt5.1

Comments URL: https://news.ycombinator.com/item?id=45982019

Points: 4

# Comments: 0

Embedding Model Leaderboard

tifa2up — Mon, 17 Nov 2025 19:24:04 +0000

Article URL: https://agentset.ai/embeddings

Comments URL: https://news.ycombinator.com/item?id=45957111

Points: 1

# Comments: 0

New comment by tifa2up in "Ask HN: What Are You Working On? (Nov 2025)"

tifa2up — Mon, 10 Nov 2025 13:32:41 +0000

Right now it's single shot, we're looking into building an "Agentic Retrieval" based on Claude ADK. tbd how it'll work

New comment by tifa2up in "Ask HN: What Are You Working On? (Nov 2025)"

tifa2up — Mon, 10 Nov 2025 07:09:06 +0000

I'm building https://github.com/agentset-ai/agentset, RAG as a service that works quite well out of the box.

We achieve this performance by baking in the best practices before any tweaking

Best Reranker for RAG: We tested the top models

tifa2up — Fri, 07 Nov 2025 20:29:42 +0000

Article URL: https://agentset.ai/blog/best-reranker

Comments URL: https://news.ycombinator.com/item?id=45850718

Points: 1

# Comments: 0

New comment by tifa2up in "Developers are choosing older AI models"

tifa2up — Wed, 05 Nov 2025 13:59:05 +0000

Think it varies by use case. It didn't do well with long context

New comment by tifa2up in "Developers are choosing older AI models"

tifa2up — Wed, 05 Nov 2025 09:14:02 +0000

For large context (up to 100K tokens in some cases). We found that GPT-5: a) has worse instruction following; doesn't follow the system prompt b) produces very long answers which resulted in a bad ux c) has 125K context window so extreme cases resulted in an error