Lawrence Xu

Building be true, be free, be helpful.

[Lawrence Xu @cdxy_]

⬇️My AI workflow, drawn by Claude Code.
analyzed workspace → pulled logos → drafted HTML → polished with gpt-image-2 via @InfronAI CLI

[Lawrence Xu @cdxy_]

Unhinged deal. $2M in credits for 2% equity. real cost to OpenAI ≈ $200K, basically 2% of a YC AI startup at $10M val. plus they probably won't even burn it. plus OpenAI owns the customer + data.

[Tyler Bosmeny @bosmeny]

A mic drop moment @ycombinator tonight

@sama just offered $2M in OpenAI tokens to EVERY YC startup in the current batch in exchange for equity

Just like Yuri Milner offering to invest in every startup back when Sam was a YC partner

I can't wait to see what's unlocked when you let the most driven, creative and formidable founders tokenmaxx

[Lawrence Xu @cdxy_]

@zechengzh wrud man?

[Lawrence Xu @cdxy_]

Promising, wanna take over MCP?

[Zecheng Zhang @zechengzh]

Introducing Mirage, a unified virtual filesystem for AI agents!

6 weeks. 1.1M+ lines of code. We rewrote bash from the ground up so cat, grep, head, and pipes work across heterogeneous services. S3, Google Drive, Slack, Gmail, GitHub, Linear, Notion, Postgres, MongoDB, SSH, and more, all mounted side-by-side as one filesystem.

Bash that AI agents already know works on every format! cat, grep, head, and wc parse .parquet, .csv, .json, .h5, even .wav! One pipe can stitch S3, Drive, GitHub, Slack, and Linear together, same Unix semantics throughout.

Workspaces are versioned too. Snapshot, clone, and roll back the whole thing with one API call. A two-layer cache turns repeated reads into local lookups, so agent loops stay fast and cheap.

Drop a Workspace into FastAPI, Express, or a browser app. Wire it into OpenAI Agents SDK, Vercel AI SDK, LangChain, Mastra, or Pi. Run it alongside Claude Code and Codex.

Site: https://t.co/zo1orc2wA9
GitHub: https://t.co/zeRAKri7I9

#AIAgents #OpenSource #AgenticAI #Strukto #Filesystem #VFS

[Lawrence Xu @cdxy_]

@zechengzh @CAISconf @struktoai @InfronAI LFG bro

[Lawrence Xu @cdxy_]

From a gateway/access layer to the soil where AI agents learn from their own data, surface insight, and optimize themselves. This is the start.

[Zecheng Zhang @zechengzh]

Excited to share that our paper, “SEAR: Schema-Based Evaluation and Routing for LLM Gateways,” was accepted to @CAISconf 2026! 🎉

This work was done at @struktoai, in partnership with @InfronAI.

Paper: https://t.co/3IjbdEe9Rz

[Lawrence Xu @cdxy_]

tbh if you're spending < $500/month on AI, you're already out.

[Elvis @elvissun]

https://t.co/DotZ3V6XhJ

[Lawrence Xu @cdxy_]

To my big tech friends:
Stop trading stocks after your 9-5.
Let's build THE stock.

[Lawrence Xu @cdxy_]

See you there!

[XerpaAI @XerpaAI]

𝗖𝗵𝗿𝗶𝘀𝘁𝗺𝗮𝘀 is coming! 🎄
Our 𝗰𝗼𝗳𝗼𝘂𝗻𝗱𝗲𝗿 @nealneal_ , experienced 𝗴𝗿𝗼𝘄𝘁𝗵 𝗲𝘅𝗽𝗲𝗿𝘁, will share how we use our own creative labs to help teams level up efficiency together with other 𝗳𝗼𝘂𝗻𝗱𝗲𝗿𝘀 @cdxy_ , @MonteCristo_ARK , @RushanLiu , @zhangboknight

See you there!

[Andrew Zheng @AndrewHan_Zheng]

Happy to grab a coffee☕

[Lawrence Xu @cdxy_]

Out here hiking in Fremont's cow valley feel like I've seen them in some movie or album before

[Lawrence Xu @cdxy_]

Happy Halloween San Francisco

[Lawrence Xu @cdxy_]

fremont 快乐老家

[Lawrence Xu @cdxy_]

Made a new website for @OneRouter
We are helping agentic startups cut their AI cost up to 30% from OpenRouter while maintain production-ready model quality, 99.98% uptime SLA and 24/7 support.
DM if you are paying OpenRouter over 5k/month, I'm ready to help👏🏻

[InfronAI @InfronAI]

🚀 OneRouter Audio Revolution is Here! 🎵

After months of development, we're thrilled to announce our Audio Model Suite - bringing enterprise-grade voice capabilities to your fingertips!

✨ What's New
🎤 5 Powerful Audio Models
• TTS-1 & TTS-1-HD: Crystal-clear text-to-speech
• GPT-4o-Mini-TTS: Advanced voice synthesis
• Whisper-1: Industry-leading transcription
• GPT-4o-Transcribe: Multi-language audio processing

⚡ Game-Changer: Unified Audio API
→ One interface for ALL audio models
→ 10x faster integration efficiency
→ Standard OpenAI format = seamless migration

📊 Enterprise-Grade Reliability
✅ Guaranteed Performance
• TPM, RPM & Latency SLAs
• Professional IT support team
• Smooth migration from OpenAI

💰 Cost Optimization Built-In
• Up to 30% API discounts for enterprises
• Default Prompt Cache = automatic savings
• B2B pre-payment options available

🎯 Real Impact
Before: Complex audio integrations taking weeks
After: Single API call handles text-to-speech, translation & transcription

"This unified approach just saved us 3 months of development time!" - Beta Enterprise User

👉 Ready to transform your audio workflow?Explore models: https://t.co/c1YAF1fnjc
docs: https://t.co/Zcgw8sdnCU

What audio use case would benefit your business most? Drop a comment below! 👇

[Lawrence Xu @cdxy_]

We also have OneRouter Enterprise - Save up to 35% on APIs from OpenRouter, OpenAI, Anthropic, XAI, Google & more

[InfronAI @InfronAI]

🚀 Generative Image & Video Models Are Now Live in Our Marketplace!

We’ve just rolled out 21+ of today’s most popular AGI models — making it easier than ever to create stunning visuals and videos with just one prompt.

✨ Highlights:
• veo3-fast → Ultra-fast text-to-video generation 👉 https://t.co/6OlWl4eqO3
• seedream → High-quality text-to-image creation 👉 https://t.co/lUx2V1qY44
• Unified API format → One prompt in, URL or base64 output out
• Advanced features → Fully documented for power users 👉 Docs for veo3-fast https://t.co/DdpkLNVeIS

📊 Real Impact:
→ One API to access all models
→ Faster integration for developers
→ Reduced time from idea to production-ready visuals

💬 “It feels like having an entire creative studio accessible via a single API call!”

👉 Explore all models now: Model Marketplace

[BagelPay @BagelPay]

Stripe is incredible — for companies.
But indie hackers ≠ companies.
The current financial infra simply doesn’t fit solo builders.
That’s why we’re building Bagelpay:
⚡️ 5-min onboarding
🌍 No company needed
💸 Global payouts
Financial infra that actually works for indie hackers. 🚀

[Lawrence Xu @cdxy_]

@tangjinzhou 加油加油！

[Lawrence Xu @cdxy_]

My co-founder hit me with "Oh it's US AGI."
I'm done with tech nerds...