june (@arjunkmrm) 's Twitter Profile
june

@arjunkmrm

vibe coding ~ human x AI interfaces, learning companions, smithery.ai

ID: 3232217942

calendar_today01-06-2015 04:31:41

1,1K Tweet

883 Takipçi

1,1K Takip Edilen

Henry Mao (@calclavia) 's Twitter Profile Photo

The future of the internet will be dominated by tool calls, not clicks. We're building Smithery to orchestrate this new era of AI-native services for AI agents. Read our mission 👇

The future of the internet will be dominated by tool calls, not clicks.

We're building Smithery to orchestrate this new era of AI-native services for AI agents.

Read our mission 👇
june (@arjunkmrm) 's Twitter Profile Photo

Context engineering is going to be a hot new field. When and how to provide the right context to a model will matter more than ever when it becomes the primary bottleneck for god like intelligence.

june (@arjunkmrm) 's Twitter Profile Photo

The biggest trap of vibe coding is your codebase becoming a blackbox to yourself. It's much harder to change your system or review changes to it if you don't have a clear mental model it.

june (@arjunkmrm) 's Twitter Profile Photo

One of my childhood friends texted me out of nowhere after finding about Smithery and told me he's been actively using it This one hit different

june (@arjunkmrm) 's Twitter Profile Photo

> run into a problem > try solving it for a few days > find a library that solves exactly that plus ones you never even imagined > your solution looks like a caveman’s sketch next to that > repeat

june (@arjunkmrm) 's Twitter Profile Photo

Claude code could be so much more than a coding tool. Coupled with MCPs, it seems like a glimpse of the universal interface from the future

june (@arjunkmrm) 's Twitter Profile Photo

Read somewhere that designing a product is like hosting guests at your home. It’s been stuck on my mind ever since

anirudh (@kamathematic) 's Twitter Profile Photo

📢 I'm cofounding Smithery with Henry Mao. In a world where no single model has all the answers, agents need a platform to access capabilities and context at scale. We're fixing this with a cohesive gateway to discover and distribute intelligence. Memo in thread 👇

📢 I'm cofounding <a href="/SmitheryDotAI/">Smithery</a> with <a href="/Calclavia/">Henry Mao</a>.

In a world where no single model has all the answers, agents need a platform to access capabilities and context at scale.

We're fixing this with a cohesive gateway to discover and distribute intelligence.

Memo in thread 👇
anirudh (@kamathematic) 's Twitter Profile Photo

The default MCP inspector is super cool, but how do you know what prompts to even send to a server? june is cooking up a prompt showcase for servers on Smithery. What else would help you trust an MCP server?

june (@arjunkmrm) 's Twitter Profile Photo

One underrated thing about using llms for coding is just helping you search stuff across huge monorepos with a ton of nested folders

anirudh (@kamathematic) 's Twitter Profile Photo

👀 I built my own minimal, functional MCP client: 👉 Next.js + AI SDK ONLY 👉 Hosted on Vercel with local storage for chat/config 👉 100% Open Source Check it out: dotcom.chat