Hacker News: samagra14

New comment by samagra14 in "Pages of Claude Mythos That Got Zero Headlines"

samagra14 — Thu, 09 Apr 2026 06:37:06 +0000

I spent the morning reading all 244 pages. About 180 of them are getting zero coverage. Here are the findings that deserve more attention from anyone building with AI esp the psych evals and p-hacking.

Pages of Claude Mythos That Got Zero Headlines

samagra14 — Thu, 09 Apr 2026 06:37:06 +0000

Article URL: https://twitter.com/samagra_sharma/status/2041882243807707230

Comments URL: https://news.ycombinator.com/item?id=47700053

Points: 10

# Comments: 1

Ask HN: What Inference Server do you use to host TTS Models?

samagra14 — Mon, 30 Jun 2025 15:50:46 +0000

All the examples I have are highly unoptimized - For eg, Modal Labs uses FastAPI - https://modal.com/docs/examples/chatterbox_tts\ BentoML also uses FastAPI like service - https://www.bentoml.com/blog/deploying-a-text-to-speech-application-with-bentoml\

Even Chatterbox TTS has a very naive example - https://github.com/resemble-ai/chatterbox\

Tritonserver docs don’t have a TTS example.

I am 100% certain that a highly optimized variant can be written with TritonServer, utilizing model concurrency and batching.

If someone has implemented a TTS service with Tritonserver or has a better inference server alternative to deploy, please help me out here. I don’t want to reinvent the wheel.

Comments URL: https://news.ycombinator.com/item?id=44424730

Points: 1

# Comments: 0

New comment by samagra14 in "Handling Unhealthy Nodes in EKS"

samagra14 — Tue, 13 May 2025 15:19:23 +0000

I am unsure as to why this is still not a native EKS feature. I understand that there can be a cascading node failure but that can be easily detected and fixed.

Karpenter version that you mention is also pretty recent. I am just surprised as to why this is not a standard thing.

New comment by samagra14 in "[dead]"

samagra14 — Thu, 24 Apr 2025 04:59:42 +0000

In this post, we’ll break down what NCCL does, why it’s critical for multi-GPU training, and how to tackle one of its common challenges – the dreaded “watchdog timeout” error.

Ask HN: Any one know good AI SRE for Kubernetes?

samagra14 — Thu, 09 Jan 2025 04:57:51 +0000

Comments URL: https://news.ycombinator.com/item?id=42641760

Points: 2

# Comments: 2

New comment by samagra14 in "Show HN: HipScript – Run CUDA in the browser with WebAssembly and WebGPU"

samagra14 — Wed, 08 Jan 2025 04:36:46 +0000

Sounds interesting! I love all these edge experiments. But as long as there is architecture dependent code for models, I feel these edge experiments can't fully express their strong suit.

You try to run something and Voila you need Ampere or Hopper or Laplace for flash attnt.

New comment by samagra14 in "Ancient Indus Valley Script Deciphered"

samagra14 — Fri, 27 Dec 2024 04:01:09 +0000

The approach is purely cryptographic and requires three base axioms

1. A guess of what type of script it is - Logopgraphic (Chinese)or Syllabic (Linear Greek) or Segmental (English/Roman).

2. A guess on what any one symbol might mean - by trial and error

3. A set of all the possible languages that might have been present in that area around that time

Linking the video here -

https://youtu.be/ncDRDSKl8uk

Ancient Indus Valley Script Deciphered

samagra14 — Thu, 26 Dec 2024 16:06:44 +0000

Article URL: https://indusscript.net/

Comments URL: https://news.ycombinator.com/item?id=42515956

Points: 8

# Comments: 2

Claude (LLMs) fails to even identify Indian Languages

samagra14 — Sat, 14 Dec 2024 05:46:52 +0000

Article URL: https://www.perplexity.ai/search/sab-kuch-mora-cheeeni-re-mose-LWjEX3zyThOW5QzXhdoyRw

Comments URL: https://news.ycombinator.com/item?id=42414937

Points: 2

# Comments: 0

New comment by samagra14 in "Developer Containers"

samagra14 — Fri, 13 Dec 2024 04:58:20 +0000

GPU workloads cannot be tested on local machines and need access to specialised hardware and specific types of GPU machines and production deployments need Docker containerised files. This abstraction connects your machine via SSH to an ec2 instance and enables hot reloading Docker containers with GPU enabled. Makes it very easy to test out containerised GPU apps.

Developer Containers

samagra14 — Fri, 13 Dec 2024 04:58:19 +0000

Article URL: https://tensorfuse.io/docs/concepts/devcontainers

Comments URL: https://news.ycombinator.com/item?id=42406031

Points: 2

# Comments: 1

New comment by samagra14 in "How to think about your documentation?"

samagra14 — Thu, 12 Dec 2024 16:31:16 +0000

At Tensorfuse we have been rewriting and restructuring our docs as they have become a huge dump of how to guides rather than being something that is navigable and understandable. After a lot of peer discussion and reading, we are now trying the diataxis approach and it seems to be working with our beta users for now. Sharing it here as some people might be looking to make their documentation more navigable.

How to think about your documentation?

samagra14 — Thu, 12 Dec 2024 16:31:16 +0000

Article URL: https://diataxis.fr/start-here/

Comments URL: https://news.ycombinator.com/item?id=42400615

Points: 1

# Comments: 1

New comment by samagra14 in "ChatGPT refuses to say the name "David Mayer," and no one knows why"

samagra14 — Sun, 01 Dec 2024 06:17:00 +0000

Perplexity with Chatgpt base doesnt break

https://x.com/samagra_sharma/status/1863104858095784094

New comment by samagra14 in "Startup Deal: Multi Cloud GPU Orchestrator Tensorfuse Free for 6 Months"

samagra14 — Thu, 10 Oct 2024 23:59:26 +0000

After being the most popular GPU deal on YC last month, we're expanding it to a wider audience.

If you’re a seed stage startup with <$3M in funding, you can run Tensorfuse free for 6 months.

Apply for the deal here and we'll reach out within 24 hrs

Startup Deal: Multi Cloud GPU Orchestrator Tensorfuse Free for 6 Months

samagra14 — Thu, 10 Oct 2024 23:59:26 +0000

Article URL: https://daily-requests-467256.framer.app/for-seed-stage-startups

Comments URL: https://news.ycombinator.com/item?id=41804798

Points: 1

# Comments: 1

New comment by samagra14 in "[dead]"

samagra14 — Sat, 28 Sep 2024 03:07:05 +0000

I’ve been working on Python-based backend development for about three years now in various forms. I primarily use Django and FastAPI, although I initially started with Flask. However, during my backend work, I frequently encountered the terms ASGI and WSGI. For example, one of my Django deployment scripts included references to asgi_app and wsgi_app, and used gunicorn to deploy these apps. Although I initially dismissed these terms as implementation details, I now find myself needing to support both ASGI and WSGI apps for my company tensorfuse. As a result, I believe it’s important to explain ASGI and WSGI to a wider audience.

Why do GPU Containers have long Cold Starts?

samagra14 — Tue, 03 Sep 2024 15:05:26 +0000

Article URL: https://tensorfuse.io/blog/gpu-containers-cold-start

Comments URL: https://news.ycombinator.com/item?id=41435667

Points: 3

# Comments: 0

WTF Is Kubernetes Autoscaling?

samagra14 — Fri, 17 May 2024 04:41:56 +0000

Article URL: https://samagra.me/wtf-is-kubernetes-autoscaling

Comments URL: https://news.ycombinator.com/item?id=40386460

Points: 2

# Comments: 0