Ask HN: Go-to places to get some ideas to work on
Asking from the Inspiration pov, like when things click
sujayk_33
Re:
on Thursday May 14, 2026 @09:59AM (#48135499)
I have poured my heart into it
Ask HN: How do you use AI agents in your day to day life?
Would love to know about workflows that you've built to make your life easier or any problem statement that you might have that you're planning to solve with an agent.
Re:Ask HN: What Are You Working On? (March 2026)
on Monday March 09, 2026 @11:58AM (#47310780)
a general-purpose mobile agent, trying out on SLMs that work on phones. In case there's no internet.
Everyday Entropy
How Ads evolved in this trendy world
Re:Ask HN: Why doesn't HN have a rec algorithm?
on Sunday February 22, 2026 @11:33AM (#47112370)
I agree too.
Ask HN: Why doesn't HN have a rec algorithm?
I was just wondering about why there's a constant timeline and no recommendation.
Ask HN: A good Model to choose in Ollama to run on Claude Code
Given that Claude Code supports a locally running model on Ollama, which is the best Thinking Model that supports tooling, can I pick for good output?Also, if anyone has tried, does it still require a Claude Subscription?(I currently have an RTX 5060 machine with 8GB of VRAM)
Tell HN: Opus 4.5 is so good that it outshadowed other LLMs
Note: Subjective to me, not picking a fight here
Re:Animated AI
on Wednesday December 31, 2025 @06:52AM (#46443390)
Yeah, it's not mobile-friendly. didn't get a chance to look into it
Re:Animated AI
on Wednesday December 31, 2025 @12:10AM (#46441509)
I worked on something similar but specifically for transformer architecture: https://transformer.sujayk.me/
Ask HN: What book are you currently reading?
I recently started reading "A Woman in Black" by Susan Hill.
A $40 Mistake
I used to read like this
Internet in the Age of Agents
Keeping the Streak Alive
Asymptotic Chase
Re:FLUX.1 Kontext
on Friday May 30, 2025 @01:26PM (#44138294)
So here is my understanding of current native image generation scenario, I might be wrong so please correct me, I'm still learning it and I'd appreaciate the help.
First time native image gen was introduced in Gemini 1.5 Flash if I'm not wrong, and then OpenAI was released for 4o which took over the internet by Ghibli Art.
We have been getting good quality images from almost all image generators like Midjourney, OpenAI and other providers, but the thing that made it special was true "multimodal" nature of it. Here's what I mean
When you used to ask chatgpt to create an image, it will rephrase that prompt and internally send that prompt to Dalle, similarly gemini would send it to Imagen which were diffusion models and they had little to know context in your next response about what's there in the previous image
In native image generation, it understands Audio, Text and even Image tokens in the same model and need not to rely on diffusion models internally, I don't think both Openai and google has released how they've trained it but my guess is that it's partially auto-regressive and diffusion but not sure about it
Re:Gemma 3n preview: Mobile-first AI
on Wednesday May 21, 2025 @03:32AM (#44049179)
https://youtu.be/eJFJRyXEHZ0
in the video they've added in announcement, they are showing some live interaction with the model(which is quite fast as compared to AI Edge gallery app), how's it built, how can I use it like this?
Re:Gemma 3n preview: Mobile-first AI
on Wednesday May 21, 2025 @03:15AM (#44049064)
8 has G3 chip
Ask HN: Best Vibe Coded projects that you have come across
I would love to see some good projects that hackers have built solely vibe-coded to see to what extent it can be stretched and to learn good practices.Anything else that you might want to share regarding the topic is also welcome.
Re:Mercury, the first commercial-scale diffusion language model
on Thursday May 01, 2025 @05:14AM (#43855291)
It's faster inference because of the Hardware (LPUs), here the question is about architectures (AR or Diffusions)
Re:
on Sunday March 02, 2025 @05:24AM (#43229072)
This is the second bullet post in the series as I prefer 10 short posts than one lengthy post.
Feedbacks are welcome.
Ask HN: How to evaluate PRs against a dynamic rubric?
I'm working on a use case where every PR created in a repository needs to be evaluated against the topic that will be tagged to it. Based on which the AI feedback is to be given, based on the feedback, changes will need to be made, until done, the merging should be blocked. when the updation of code happens, it should be able to check it against the previously given feedback. Now it can ask for...
Ask HN: Free/Open Source Brainstorming Agents
Are you aware of any good Brainstorming agents I can converse with? Gemini has one but needs to have the Gemini Advanced for that.Normal LLM also works out but would love to have a specialized AI for that. I love brainstorming with AI.
Ask HN: What's the craziest thing a copilot did for you?
Just wonderin
Language Models: A Guide for the Perplexed (2023)
A Deep Dive into Memorization in Deep Learning
Re:Ask HN: What is your definition of luxury?
on Tuesday January 14, 2025 @09:45AM (#42697800)
Probably the most difficult one to obtain