Hacker News: applgo443

New comment by applgo443 in "Mistral OCR"

applgo443 — Fri, 07 Mar 2025 00:23:27 +0000

What's the simple explanation for why these VLM OCRs hallucinate but previous version of OCRs don't?

New comment by applgo443 in "Ingesting PDFs and why Gemini 2.0 changes everything"

applgo443 — Thu, 06 Feb 2025 06:48:30 +0000

Why are traditional OCRs better in terms of hallucination and confidence scores?

Can we use logprobs of LLM as confidence scores?

New comment by applgo443 in "Show HN: Envelope – A modern environment variable cli tool"

applgo443 — Sun, 23 Jun 2024 00:43:26 +0000

When i deploy a webapp on azure, it expects me to put env variables in a file or their own tool (key value fields) where you can add env variables one by one.

Is there a way to use envelope in places like those?

New comment by applgo443 in "RedPajama v2 Open Dataset with 30T Tokens for Training LLMs"

applgo443 — Tue, 31 Oct 2023 05:15:12 +0000

If it's 5 common crawls, isn't data across multiple common crawls mostly similar?

New comment by applgo443 in "Launch HN: Serra (YC S23) – Open-core, Python-based dbt alternative"

applgo443 — Tue, 15 Aug 2023 00:14:06 +0000

Is ETL/ELT same as writing SQL scripts and periodically executing them? I assumed there's more to it.

New comment by applgo443 in "Launch HN: Sweep (YC S23) – A bot to create simple PRs in your codebase"

applgo443 — Fri, 04 Aug 2023 03:34:01 +0000

May be rewriting user's description might help you match code better?

Similar to the prompt engineering for previous era GPT completion models.

New comment by applgo443 in "Launch HN: Sweep (YC S23) – A bot to create simple PRs in your codebase"

applgo443 — Fri, 04 Aug 2023 03:32:27 +0000

How do you approach the problem of what files to look into to fix a bug? Just embeddings doesn't seem to cut it.

New comment by applgo443 in "Launch HN: Sweep (YC S23) – A bot to create simple PRs in your codebase"

applgo443 — Thu, 03 Aug 2023 22:20:43 +0000

How is your experience with Modal?

And I'm curious to know more about your costs of deployment and running on Modal.

New comment by applgo443 in "Launch HN: Sweep (YC S23) – A bot to create simple PRs in your codebase"

applgo443 — Thu, 03 Aug 2023 22:19:26 +0000

Did you consider first asking LLM to explain what a code snippet does and use that instead?

It'd significantly increase the costs though.

New comment by applgo443 in "Langchain Is Pointless"

applgo443 — Sat, 08 Jul 2023 22:03:21 +0000

I saw your comment, got curious, and looked at a lot of your old comments. Lots of interesting insights - Thanks for sharing them.

If you don't mind me asking, what do you do? I'm a researcher at FAANG working on language models and starting a new company in the space. Would love to connect. Feel free to email me - idyllic.bilges0p@icloud.com

LLM Training Puzzles

applgo443 — Fri, 30 Jun 2023 23:57:24 +0000

Article URL: https://github.com/srush/LLM-Training-Puzzles

Comments URL: https://news.ycombinator.com/item?id=36544355

Points: 1

# Comments: 0

New comment by applgo443 in "ChatHN: Chat with Hacker News using OpenAI function calling"

applgo443 — Mon, 26 Jun 2023 17:07:09 +0000

Used OpenAI APIs before but not function calling. What's the use of function calling in this example?

New comment by applgo443 in "Show HN: Autolabel, a Python library to label and enrich text data with LLMs"

applgo443 — Wed, 21 Jun 2023 21:03:16 +0000

How does the confidence scores work?

New comment by applgo443 in "Emerging architectures for LLM applications"

applgo443 — Wed, 21 Jun 2023 20:32:18 +0000

I’m a researcher in the space exploring few ideas with the intention of starting up. Would love to reach out to you and talk to you. Is there a way I can contact you?

My email is beady.chap-0f@icloud.com

New comment by applgo443 in "Emerging architectures for LLM applications"

applgo443 — Wed, 21 Jun 2023 00:07:30 +0000

I think we can detect atleast a few things like PII leaks etc. Don't you think those things alone are valuable?

New comment by applgo443 in "Emerging architectures for LLM applications"

applgo443 — Tue, 20 Jun 2023 20:45:09 +0000

They mention the contextual stack is is relatively underdeveloped. Any idea on what can be improved there?

New comment by applgo443 in "Emerging architectures for LLM applications"

applgo443 — Tue, 20 Jun 2023 20:17:06 +0000

What do you mean by firewall layer? What tools do you use here?

Ask HN: What does it take to be a CTO?

applgo443 — Sat, 17 Jun 2023 20:04:11 +0000

Hi,

A friend of mine and I are thinking of starting a company together. My friend is a sales/marketing person (CEO), and I'm the (only) technical person (CTO).

I want to know what it really takes to be a CTO. On one end, I see people starting companies directly after their college as a CTO. On the other end, I see people who were architects at FAANG, built massive open-source projects before starting their companies as CTO. I'm somewhere in between - I have a few years of experience at FAANG as a Senior ML Engineer, but most of my work has been on platform/ML research rather than things like full stack development, deployment, etc.

I am trying to figure out if I'm capable enough to be a CTO. I understand there will be a lot of on-the-job learning, but my question is more about how much I should already know on day one.

The question is deliberately a little vague as it's tough to convey my full background. I would appreciate any tips or suggestions here.

Comments URL: https://news.ycombinator.com/item?id=36374023

Points: 25

# Comments: 17

New comment by applgo443 in "Inflection AI launches pi chatbot"

applgo443 — Tue, 02 May 2023 19:56:20 +0000

Wow, if you're competing with ChatGPT, that gives all other startups out there hope that competitors aren't something to be super scared about :)

New comment by applgo443 in "Ask HN: Who wants to be hired? (April 2023)"

applgo443 — Mon, 03 Apr 2023 23:52:21 +0000

Location: SF Bay Area

Remote: Preferred

Willing to relocate: Depends on opportunity

Technologies: Python, Pytorch, Tensorflow, Jax, machine learning, deep learning, language modeling,

Email: beady.chap-0f@icloud.com

I'm a Research Engineer at Apple, working on Language Models. Have expertise in both building LMs + using these LM APIs to build apps. I'm adept at the fundamentals of machine learning, deep learning and worked across NLP (language models), NLP + CV (Image captioning), and reinforcement learning (PPO for image captioning, robot navigation, etc). I am also a proficient C++ systems developer and worked on several low level systems problems such as sorting, caching, etc before I entered ML space.

What I'm looking for: Though I'm primarily a research engineer, I'm interested in joining a place that lets me wear multiple hats, and gives me enough responsibility and opportunity to scale my personal learnings and skills along with that of the company. My mid term goal (~ 5 years) is to start my own company. Hence, for my next opportunity, I'd like to join a small startup where I can also learn the non-technical parts of building a company.