Hacker News: iAkashPaul

New comment by iAkashPaul in "No GPS required: our app can now locate underground trains"

iAkashPaul — Wed, 13 Nov 2024 12:11:50 +0000

Could be that or an RNN/LSTM as well

New comment by iAkashPaul in "Claude AI to process secret government data through new Palantir deal"

iAkashPaul — Sat, 09 Nov 2024 00:35:50 +0000

Reversed input strings will the next Enigma

New comment by iAkashPaul in "More annoying macOS 15 Sequoia prompts: Bluetooth"

iAkashPaul — Sun, 29 Sep 2024 14:24:31 +0000

IMO hitting CMD+T thinking it'll open a new tab in my browser but it opens the font window (from Mail) is a user-crime.

Building an open-source full-frame CCD camera PART II (2024) [video]

iAkashPaul — Sat, 28 Sep 2024 16:17:09 +0000

Article URL: https://www.youtube.com/watch?v=OkfzjmY9cF8

Comments URL: https://news.ycombinator.com/item?id=41681167

Points: 3

# Comments: 1

New comment by iAkashPaul in "The Cheating Device (ChatGPT on a TI-84) [video]"

iAkashPaul — Thu, 19 Sep 2024 17:37:05 +0000

A USB MITM board with ESP32 could just connect to your phone & do one-way code/content creation for otherwise software locked devices.

New comment by iAkashPaul in "Set Up a $4/Mo Hetzner VM to Skip the Serverless Tax"

iAkashPaul — Mon, 02 Sep 2024 05:48:49 +0000

Scaling/cold-boot time is the real concern IMO no the cost

New comment by iAkashPaul in "Benchmarks show even an old Nvidia RTX 3090 is enough to serve LLMs to thousands"

iAkashPaul — Sat, 24 Aug 2024 17:54:18 +0000

Pretty sure this was never questioned for batched requests, sg-lang/lmdeploy/tensorRT-LLM will have nearly twice as reported speeds with INT8 (fp16 A100 benched here https://github.com/sgl-project/sglang?tab=readme-ov-file#ben...)

New comment by iAkashPaul in "Ask HN: Visualize Software Architecture/Concepts"

iAkashPaul — Mon, 12 Aug 2024 06:43:53 +0000

Apart from draw.io I like to use isoflow.io for near isometric diagrams with directional flows as well

New comment by iAkashPaul in "Show HN: Pie Menu – a radial menu for macOS"

iAkashPaul — Mon, 05 Aug 2024 19:11:37 +0000

Hey Marius, cool project! Would this work with just the middle mouse button click too?

New comment by iAkashPaul in "Taking command of the Context Menu in macOS"

iAkashPaul — Wed, 31 Jul 2024 10:55:04 +0000

More than this pressing spacebar should have gotten some customisation by now, eg. certain mp4s/markdowns/any other file-extension for quick previews

New comment by iAkashPaul in "Intel N100 Radxa X4 First Thoughts"

iAkashPaul — Sun, 28 Jul 2024 11:23:02 +0000

Can't wait to get this & have TrueNAS+Jellyfin up & running on it, certainly beats having to buy so many addons for the Pi & still not have a super reliable build.

New comment by iAkashPaul in "Ask HN: How is your organization using LLMs?"

iAkashPaul — Fri, 26 Jul 2024 07:59:49 +0000

Where did you find OpenAI in my reply?

New comment by iAkashPaul in "Glacier is an open-source web browser with extensive customization capabilities"

iAkashPaul — Thu, 25 Jul 2024 08:39:17 +0000

Thx will check it out

New comment by iAkashPaul in "Solving the out-of-context chunk problem for RAG"

iAkashPaul — Thu, 25 Jul 2024 05:53:26 +0000

One quick way to improve results greatly is to ask questions with 2/3 chunks & in the lookup for these chunks mention the IDs of the other chunks, qdrant allows for easy metadata addition. So just generate a synthetic question bank & then do vSearch against the same instead of hoping for the chunks to match up with user questions.

New comment by iAkashPaul in "Glacier is an open-source web browser with extensive customization capabilities"

iAkashPaul — Wed, 24 Jul 2024 06:03:59 +0000

What I wished browsers on Android would do is scroll pages with volume buttons and close pages on double tap or some combination similar to screenshot ring.

New comment by iAkashPaul in "Ask HN: How is your organization using LLMs?"

iAkashPaul — Fri, 19 Jul 2024 14:56:19 +0000

Mostly Mistral/Mixtral models in addition to Whisper.

New comment by iAkashPaul in "Ask HN: How is your organization using LLMs?"

iAkashPaul — Fri, 19 Jul 2024 08:41:41 +0000

Heavily,

Call summarisation at scale, for further insights of key topics/sentiments for diarized & fused channel call data

Lead mining quality improvement by asking questions instead of merely classification

Internal simple RAG chatbot across ~1000 PDFs, not yet down the GraphRAG route

Better chatbot experience over RASA

Contract mining for speeding up mundane governance tasks

New comment by iAkashPaul in "Benchmarking LLM Inference Back Ends: VLLM, LMDeploy, MLC-LLM, TensorRT-LLM, TGI"

iAkashPaul — Sun, 07 Jul 2024 02:30:37 +0000

Nice, I'll bench server.cpp as well

New comment by iAkashPaul in "The Framework Laptop 13 is about to become one of the first RISC-V laptops"

iAkashPaul — Wed, 19 Jun 2024 08:18:40 +0000

How long until someone makes a harvested Apple Silicon logic board work with this!

New comment by iAkashPaul in "Every Way to Get Structured Output from LLMs"

iAkashPaul — Tue, 18 Jun 2024 08:05:54 +0000

My point exactly, lot of brigading on this thread