Soloman Sunnie (@NbSoloman) header
Soloman Sunnie (@NbSoloman) profile picture

AI Builder · Indie Hacker Building with AI | Agents | Web3 Vibe Coding · Shipping Products · Building in Public

Singapore
Joined October 2019
Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)Photo by Soloman Sunnie (@NbSoloman)
Want to grow on X? Stop lurking. Start connecting. AI. Web3. Tech. Builders. If that's your world, let's connect. 🤝
6
4
179
Amazing 🥳
📢 Meet Qwen4-Max, your new AI teammate for work, creativity, and discovery! Qwen4-Max is built on a new MoE architecture with native vision and a 1M-token context window. Highlights: - Outperforms Qwen3.8-Max with up to ~50% fewer tokens, while average cost per task drops over 60% in our internal evaluation. - Advances real-world productivity through greater proficiency with professional tools, stronger long-horizon execution, and higher-quality deliverables. - Handles real tasks, including 3D modeling, game development, manufacturing, finance, law, healthcare, and frontier scientific discovery. - $1.5/M tokens for input, $4.5/M tokens for output, and $0.075/M tokens for implicit caching — all lower than Qwen3.8-Max, with implicit caching down by 70%. The API is live on QwenCloud. You can also try Qwen4-Max on Qwen Studio, QwenWork and Qoder. Start building today! 🚀 - Blog: qwen.ai/blog?id=qwen4-max - API: qwencloud.com/models/qwen4-m… - Token Plan: qwencloud.com/pricing/token-… More to come! The open weights of smaller Qwen4 models will be released. Stay tuned. 👀
2
64
My feed needs MORE people who actually build things. 🚀 I'm looking to connect with people who are: 🚀 Building startups & side projects 🤖 Exploring AI, agents & new tools 💻 Coding, designing & shipping products 📈 Experimenting with Web3 & crypto ✍️ Mastering marketing, sales & growth Building something? Working on a crazy idea? Just obsessed with tech? Let's connect. What you're building below! 👇
20
22
763
Want more from AI? Stop handing it tasks. 1. Aim at the outcome, not the task 2. Go as deep as a real expert would 3. Borrow methods from other fields 4. Validate with something outside the model, and give it real tools to work with Ambition is the new skill. felixrieseberg.com/how-to-be…
1
3
104
WTF?! 🤯 You can literally just press and hold any GIF, copy the link, paste it, and BOOM — it's fucking posted. How the hell did I not know this before?! 💀
4
377
Hi @X Looking to #connect with people who love building things and exploring what’s next. 👀 👨‍💻 Developers & Indie Hackers 🤖 AI Enthusiasts & Builders 🌐 Web3 Explorers 🎨 Designers & Creators 🚀 Founders & Tech Lovers If you're building, experimenting, or simply passionate about technology, let's connect! What you're working on. 👇 Let's grow together! 🤝
27
1
27
673
Your AI Agent might be wasting 42x more tokens just to use a tool. MCP vs CLI vs Skills — OpenRouter put them to the test. On a GitHub task: CLI: 1,421 tokens Skills: 3,315 tokens MCP (46 tools): 59,417 tokens Use CLI when possible. Add Skills to teach agents how to use custom tools. Use MCP when you need structured interfaces, remote access, and permissions. And don't forget: tool outputs can be just as expensive as tool definitions. The future of Agent engineering isn't just about giving agents more tools. It's about giving them the right tools with the least context overhead. openrouter.ai/blog/insights/…
5
6
144
Official Statement: I am NOT a Chinese spy 💃
3
5
871
AI Agents are more than just LLMs calling tools. There's a whole architecture behind them. 🤖 I'm looking to #connect with people exploring: Agent Loops, Context, State & Memory Tools, Skills & Protocols Runtime & Sandbox Observability & Evaluation Security & Governance MCP, Plugins, Subagents & Hooks External Session Stores Building, breaking, or experimenting with AI Agents? Let's connect and exchange ideas! 🚀
9
5
274
ByteDance Seed just unveiled Seedream 5.5, its next-generation image generation model. Three major upgrades: 1️⃣Pixel-perfect preservation: Keeps untouched areas intact, with up to 100 consecutive editing rounds on the same image. 2️⃣Native RGBA generation: Creates transparent assets with an Alpha channel built in. No more manual background removal. 3️⃣Enhanced visual quality: More realistic lighting, materials, and skin textures, alongside significantly improved text rendering.
2
7
308
AI Agents are changing how we build software. Build what you wish existed. Make it yours. Ask for what you want. Let AI build it for you. Stop tweaking. Start using. The best software might be the software you build for yourself.
1
92
Hey everyone! 👋 I'm an AI developer from Hong Kong, passionate about AI and emerging technologies. If you're into AI and tech, let's connect! 🤝
70
68
2,495
AI agents can do the work. But are we slowly forgetting how to think? The real danger isn't AI making mistakes. It's humans losing the ability to recognize them. Delegate execution, not judgment. Automate the work, not accountability. Think critically. Verify with evidence. Own the outcome. AI can take the work off your hands, not the responsibility.
2
4
219
Wow, this offers a new approach to model training — using a powerful model to edit and optimize the weights of a smaller one.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math. rikiyat.github.io/blog/weight-sp… github.com/RikiyaT/lm_und… huggingface.co/Qwen/Qwen3.5-0… huggingface.co/RikiyaT/Qwen3.…
1
149
Attention Is All You Need. Agents Are More Than LLMs. Agent engineering is about the entire stack: Agent Loop, Context, State, Memory, Tools, Skills, Protocols, Runtime, Sandbox, Observability, Evaluation, Security & Governance. MCP, Plugins, Subagents, Hooks, and external Session Stores bring it all together. The future is agent architecture.
3
193
Every day is Zeroday. ⚠️ Wiz researchers uncovered at least 9 security incidents across AI developer platforms in October alone. Fireworks AI, Lightning AI, Railway, Warp, OpenCode… the list keeps growing. Credential theft. Agent swarm attacks. Model theft disguised as distillation. The AI boom is creating a massive new attack surface. Build fast. But never forget security. 🔐 wiz.io/blog/black-october-fo…
1
3
194
Vibe coders, what have you been building? 👀 Show us what you've created! Drop your projects in the replies 👇 Let's discover cool ideas, share inspiration, and connect with fellow builders! 🚀
11
12
465
🚀 WTF?! Gemini is suddenly cracked.
Gemini 4 Carbon is our next frontier model. It’s state-of-the-art across cybersecurity and long-horizon capabilities.
1
104