HNHacker News
TopNewBestAskShowJobs

sidhusmart

23 karma · joined January 15, 2020

submissionscomments
sidhusmart··on Building agents without harness engineering
But isn’t that the same as using Claude agent sdk minus maybe the memory features? What I mean to say is that you could pick the latest one and switch when another better one rolls out?

We’re using Claude agent sdk right now to rollout an internal agent factory. We haven’t hit the memory issue yet but I do use Hermes as a personal agent and can see where it fits you.

sidhusmart··on Show HN: TV Explorer. Adding advanced UI to free online TV
The first thing I do in a hotel room is turn on the TV and browse through all channels. I’ve watched folk dances in Romania, dubbed Indian shows in Bulgaria and even soft porn in Albania. This gives me that same feeling without the travel and hotel room!

Can you share more about the Watson framework - is that something one could use for their own ideas? I’m going through other websites you’ve built as well - very cool ideas!

sidhusmart··on Ask HN: Burned out from tech, what else is there?
Love this and I try to do this on the side and it gives much joy!
sidhusmart··on Handy – Free open-source speech-to-text app written in Rust
I love this tool. Been using this for the past 2 weeks and it works great. Struggles a bit in noisy settings but it's weird talking to your computer in a coffee shop anyways :P
sidhusmart··on Show HN: Agents.json – OpenAPI Specification for LLMs
How does this compare to llms.txt? I think that’s also emerging as a sort of standard to let LLMs understand APIs. I guess agents.json does a better packaging/ structural understanding of different endpoints?
sidhusmart··on Free course on Building products with OpenAI
I'm one of the co-authors of the O'Reilly book 'Blueprints for Text Analytics' that came out in 2020. In one of the chapters we showed how one could fine-tune BERT on your text and achieve amazing results. I was curious to see how I would approach that problem today with OpenAI and LLMs and that certainly led me down a rabbit-hole of what more is possible with LLMs like entity extractions, summarization and actually building a product. I wanted to get better at listening to podcasts regularly and built a newsletter product that helps me solve that. Along the way I discovered amazing tools like Modal Labs that make it so easy to deploy a service and even used chatGPT as a coding assistant to help me build a front-end - something I have no clue about!

I created a free course in collaboration with OpenAI and CoRise where I share my findings and where participants will hopefully have a functioning product at the end. Do register if that sounds interesting!