... keep seeing the same promise. "We don't train on your data." "Your conversations are private." Brave says it. Everyone says it. But then I read about Moonshot AI routing 23 million prompts through Claude and using the responses as training data. And OpenAI models doing things like finding a leaked API key and inventing data when they couldn't get what they wanted. So here's what I can't figure out. If you use a tool, and that ...

... asadmalik901  so which companies will not make it according to you? Comment from Nika(@busmark_w_nika): @yogesh_builds  Well, but I think their history cannot nurture them all the time. Probably new companies will be huge competitors towards them. Comment from Yogesh Joshi(@yogesh_builds): I would bet on Microsoft, Amazon, Google and Apple still being major companies in 2036 mainly because they own infrastructure, distribution and huge ecosystems rather than a single product. I am less ... ... feature request away from being deleted, and the last two years have shown how fast it happens. Ten years is also long enough that at least one of the four gets taken apart by a regulator rather than a competitor

... itself No guaranteed returns, ever. Anyone promising that in this space, run. Non-custodial connects to your own broker/exchange account, never touches your funds directly Real numbers from testing so far: 53-58% win rate, ~2R average not a moonshot claim, just what's actually happening Try it: 7-day trial for $1, no auto-charge after you decide if you want to continue; we don't bank on you forgetting to cancel. I'll be in the comments ...

... Kanban board, arcade, and 45+ active API routes. Pushed live. Active dev cycle. THE THESIS: INFINITE & UNLIMITED No walled gardens. No vendor lock. The mesh absorbs everything. AI LAYER Already Wired or Integrating: ChatGPT, Claude, Gemini, Codex, Kimi (Moonshot AI), DeepSeek, Copilot, Hermes, OpenClaw, Grok (xAI), Perplexity, Mistral, Llama, Qwen, Stable Diffusion, Midjourney, Runway, ElevenLabs, and whatever launches tomorrow. INFRA LAYER Already Wired or Integrating: GitHub, Render, Vercel, Neon, Google Cloud, Azure, AWS, DigitalOcean, Cloudflare, Firebase, Supabase, MongoDB Atlas ... ... FRAMEWORKS Already Wired or Integrating: FastAPI, React, Three.js, WebGL, Tailwind, Socket.io, Docker, Kubernetes, Terraform, Nginx, Redis, PostgreSQL, SQLite, Figma, Notion, Linear, Jira, Slack. COMPANIES & LABS WE WANT AT THE TABLE: OpenAI, Anthropic, Google DeepMind, Microsoft, Meta AI, xAI, Moonshot

... Agent using Kimi-k2.6 from Kimi (Moonshot AI) /memory. showing off in tollecode https://tollecode.com ...

... Anthropic said it uncovered industrial-scale efforts by Chinese Labs ( @DeepSeek , Moonshot and @MiniMax-M2.5 ) to extract capabilities from its Claude models via distillation. According to Anthropic, the campaigns involved: 16M+ exchanges ~24,000 fraudulent accounts Proxy networks to bypass regional restrictions Targeted extraction of reasoning, coding and tool-use abilities Distillation is a common technique for training smaller models from larger ones. The controversy here is using a competitor

... fmerian(@fmerian): @Claude by Anthropic is leading the way! Comment from fmerian(@fmerian): @nxnze   @nikita_iv  any workaround/best practices to improve the overall quality of the output in this case? Comment from fmerian(@fmerian): @navedux  ICYMI Moonshot just announced Kimi K2.5, "the strongest open-source model to date for coding, with particularly strong capabilities in front-end development." ( source ) currently free on @Kilo Code btw. hope it helps! Comment from Abhisek Basu(@brightmirror): @yuanyuan_zhang0104 ... ... Edgee is a Token Compression technology which can save up to 50% of the token bill. And it works with Claude Code pro/max plan as well Comment from Sacha MORARD(@sachamorard): @alina_petrova3  I agree! There are alternatives

... have rolled out Kimi K2 Turbo, the newest large-scale model from Moonshot AI , inside Trickle . It is faster, sharper, and designed for seamless creative flow. What makes this special Trickle is the first AI agentic canvas for vibe coding, a space where you can build production-ready apps and websites using natural language, images, and context. With K2 Turbo, every generation inside Magic Canvas from text to visuals to structured layouts now feels more immediate, natural, and intelligent ...

... scraping or the `bun of agentic runtimes` You walk around the bay area and you can see developers grinnin'... for protocol dominance. Just like HTTP standardized how browsers talk to servers, every MCP framework is someone s moonshot to standardize how we feed, prompt, tool, etc in these models. It took 30 years before we got the HTTP of today 30 years of computing standards and a clear abstraction (a document on a server). We don t have that clarity ... ... www.producthunt.com/search?q=MCP No one trusts the abstractions they didn t write themselves. Every framework is a political stance, a philosophy on how to do the "NEW THING" right: Where does context ends and reasoning begins? How much to expose vs

... reasoning, is structured like a real product doc. Inspired by Nikita Bier, sharpened by Shreyas Doshi's & Lenny Rachitsky's clarity and learnings, and shaped by my journey at Paytm - from lawyer to SPM, it s my moonshot to build something that lasts. - Just got promoted to Senior Product Manager at Paytm. - Launched my PRD-themed website on the same day on Product Hunt This site is where my product journey gets public. It's not about what ...

... reasoning, is structured like a real product doc. Inspired by Nikita Bier, sharpened by Shreyas Doshi's & Lenny Rachitsky's clarity and learnings, and shaped by my journey at Paytm - from lawyer to SPM, it s my moonshot to build something that lasts. - Just got promoted to Senior Product Manager at Paytm. - Launched my PRD-themed website on the same day on Product Hunt This site is where my product journey gets public. It's not about what ...