Attributed mentions
164
Across all collected Hacker News results
Discussion intelligence
A source-linked Hacker News monitor for the tracked company graph. These are attributable public discussions—not sentiment, endorsements, or unverified company facts.
Attributed mentions
164
Across all collected Hacker News results
Companies represented
12
Exact-name monitor matches only
Collection source
Hacker News
Latest new record seen Today
Source-backed records
Showing 81–100 of 164 matching discussions
>I don’t care whether you wrote every word with your bare hands. I care whether the thought is yours, whether you exercised judgment, whether the result deserves my attention and whether you’ll stand behind it once it leaves your screen. Alright, compare the following degrees of autonomy: - I wrote it, made sure everything's correct, let the LLM proofread it/translate it to English, and checked it afterwards - I had brief scattered notes with no coherent idea, then threw it at an LLM to make it an article in whatever way it feels best because I don't care - I had a three word prompt "Write about X", then a deep research agent spent ungodly amount of tokens and search and scraping requests, and put up an article (at least some grounding and it can depend on the harness quality) - I…
I suspect certain workflows like langchain and others like it will retain usefulness into the future. Having deterministic steps before and after the llm is the way to go for anything that might be potentially harmful. Which I guess is what the harness does, but why be limited to a generic harness when we can use it to make specific ones for our needs.
Great project! have you considered giving the agents code-based video rendering like MoviePy or Remotion? Since LLMs are mostly trained on texts and code maybe for edits use code based approach and for complex scenes fallback for seedance, elevenlabs mcp or connectors
I would rather we not lose the AI race so idiotically. Winning the AI race means letting OpenAI and Anthropic face market forces. Edit: Clearly other companies are doing well. ElevenLabs, Runway, FAL, RunPod, CoreWeave, Lambda, BaseTen, Lovable. These are real companies that didn't raise absurd amounts of capital. They've grown revenues at a solid clip, provide real value, and didn't go off the deep end with fundraising.
By what metric is Mistral competitive? And ElevenLabs is incorporated in the USA
Mistral and ElevenLabs seem to be doing pretty well. Also, here in the U.S., we’ve managed to screw up so badly on data privacy and basic rights that the EU has been systematically replacing their entire tech stack with sovereign components.
Yes, I do semantic chunk to better split the text into coherent chunks. I even use Llamaparse api (by Llamaindex) to parsing any type of documents, images or imported study material, with a very high quality results. Llamaparse is great!
Nice use of Qdrant for the retrieval layer.Curious how you're handling the question generation, are you chunking the study material and generating per chunk, or doing something more document-level?
After trying out various study tools, I decided to create my own. I wanted a quick and simple way to upload any study material, have the app generate questions, and then set up a daily review schedule. So, I built it by creating a RAG system and a scheduler to trigger the daily review sessions.
I have. They are not to my taste. Here's a voice I really enjoy listening to. https://elevenlabs.io/app/voice-library?search=zNsotODqUhvbJ... If I could have Christopher Lee, or Stephen Fry, I would.
I've been working on this Claude Code pipeline for making 10-minute movies, gluing together Seedance (videogen), Nano Banana (image gen), and ElevenLabs (voice gen), and using Claude as the director. My repo contributes a markdown playbook and a worked example which should make it easier for other people to make their own short movies. A first-pass movie takes around 2.5 hours of wall clock time and costs around $200, and iterating is easy thanks to intermediate artifacts like storyboards and voice samples.
I made a little python wrapper around it to serve an HTTP endpoint that’s OpenAI/elevenlabs compatible https://github.com/clayrosenthal/bootlegger
Wouldn't most opensource harnesses work with lmstudio? I assume it has an "openai" style chat API like every other model provider? What's special about it vs langchain deep agents or pi or pydantic-ai?
I've been enjoying the Michael Caine voice clone narration of the Odyssey, but noticed it uses Greek names (Odysseus, Poseidon, Zeus), whereas the William Cullen Bryant translation ElevenLabs used as the basis for the audiobook uses the Roman names (Ulysses, Neptune, Jove). I tried reading the WCB translation while listening, but it got so confusing I had a go at making a companion edition that matches the names in the narration. It's free, non-commercial, and obviously not an official ElevenLabs product, just a fan attempt. Mostly for my own use, but thought I'd share in case it helps anyone else reading The Odyssey ahead of the Nolan film :)
I have developed an open-source memory system for agents accessible over MCP, which makes it possible to access it via any coding agent locally (claude-code) or via mobile (Claude AI, Mistral AI, etc). The primary storage mechanism is .md files stored as Notes in Nextcloud; however, since Nextcloud supports a rich ecosystem of apps such as documents, rss feeds, calendar, etc, your knowledge base can grow with you. All content can optionally be indexed and available via semantic search - powered by a Qdrant vectordb. The biggest cost drivers of a system like this is the memory required to host the vectordb - I'm really curious how others are optimizing their knowledge base. Thanks OP for doing to work in summarizing these tools! If you're interested either the MCP server or Nextcloud App f…
Full-Stack Developer | India (Open to Relocation & Visa Sponsorship) Tech Stack: React, Next.js, TypeScript, Node.js, FastAPI, PostgreSQL, MongoDB, Docker, GitHub Actions, REST APIs, AI/LLMs (LangChain, OpenAI, Gemini) Open-source contributor to Meshery (CNCF), KubeEdge (CNCF), and omegaUp. Built production-ready projects including Error404 UI and InsightCareer AI. GitHub: https://github.com/manishpatel00 � Portfolio: https://manishdevin.vercel.app Error404 UI: https://error404-ui.vercel.app Resume: https://drive.google.com/file/d/1g1feAoFWI7AVZBjEM0UYAFBpbr1... I'm interested in the Founding Full-Stack Software Engineer role and will also send my application via email. Thank you!
I addressed this a little bit in the comment below, but the cycles add up. I'm doing some pretty crazy things higher up in the stack, that I'm not quite ready to release yet. But even micro optimizations here add up at the scale I'm working at to allow me the headroom I need. I'm relying on a high-frequency recursive agentic loop that chokes a real-time guarantee without every optimization I can give it. I started by removing the IPC overhead from a weaviate db connection, and doing all my vector math in house with a lightweight sqlite db. This became the next obvious target for optimization once I saw how much that saved me doing things in-house.
I get what your saying, however we don't need to be stuck with the Claude code harness. You may find this interesting. https://www.langchain.com/blog/tuning-the-harness-not-the-mo...
Instrumentl | Senior (and above) Backend Engineer | REMOTE (US + Canada) | Full-time | 175,000 - 220,000 | Instrumentl is a profitable, hypergrowth, YC-backed SaaS platform building the operating system for grant-funded organizations. More than 5,500 nonprofits use Instrumentl to discover, track, and win grant funding, from local community organizations to the San Diego Zoo and the University of Alaska. Collectively they’ve moved over $1 billion through our platform. Looking for a strong Python engineer to build AI features end to end, from rapid prototype to production and the evaluation that keeps them honest. Join the engineering of 22 and company of almost 100 located all over the world. Note, AI experience is a plus, but not required. Stack: Python (FastAPI), Langchain, Postgres, Red…
Methodology: HN Search returns recent public items matching a monitored company name. AIIStack stores a short normalized excerpt and the original link, deduplicates by company/provider/item ID, and creates an activity signal only after a threshold of newly observed records. Review the original discussion before making a decision.