akashtndn
Loose opinions on future of software, developers and corporations
Abundant Software, Uncertain Economics
Show HN: Ek Din – interactive website to experience Indian 90s nostalgia
I built this website after some recent viral Astra/Opus generated content reminded me of an old side project idea I'd given up on.The content should be particularly nostalgic and hopefully comforting for people who grew up in India during the 1990s. But even beyond that, I think it is an interesting showcase of how feasible certain kinds of projects have become using coding agents.Whether you...
The Craft of Software Scoping
Embedding OpenAI Codex: The App Server and SDKs
Stagehand: The AI-Driven Browser Automation Framework
Show HN: Explain this – agent skills to create HTML explainer for topics or code
https://github.com/analyticalmonk/explain-this/
Over the last few months, I have frequently created HTML explainers using coding agents when trying to understand new topics or exploring codebases. Nothing fancu but the approach has been helpful. Some examples are here: https://www.akashtandon.in/interactive-explainers.After being asked by a couple of friends to share the workflow, I decided to create a skill plugin for the same. Hopefully,...
Turning Strava data and gym photos into a training recap with my coding agent
Debugging in the Age of Agents
Fine-tune FLUX.2 [Klein] with a LoRA under 60 minutes
Superpowers: The Anatomy of an Agent Skill
Autotrader – paper trading AI agent for Indian equities
Awesome Neuroscience
Autotrader: An autonomous paper trading agent, two weeks in
Artemis II: Why Going Back to the Moon Is a Big Deal
DSPy: Programming – Not Prompting – Language Models
Firestore Schema Visualizer
Interactive Explainer: Pi and OpenClaw
Fal-AI-skill: Claude Code plugin for working with fal AI
Four Years a Founder: Time
A 7-Step Guide for doing UX research for AI features
Lessons from building an AI feature that works – chunking and summarization
Re:What's the best explanation for LLMs being almost right when citing...
on Sunday June 09, 2024 @02:35PM (#40626484)
The prompt isn't relevant to this question though. The quality of output can be improved with better input but in this case, I am curious about the underlying mechanics in the model that leads to such behavior.
Firestore schema visualizer: Schema visualization using GPT-4 and PlantUML
What's the best explanation for LLMs being almost right when citing sources?
For example, I asked GPT-4 to identify the source for a quote from a short story by Edgar Allan Poe. It correctly identified the author but not the title. What's the best available explanation for such behavior by LLMs?