dylibso

909 posts

dylibso banner
dylibso

dylibso

@dylibso

Pioneering AI-driven app extensibility.

San Francisco, CA Katılım Temmuz 2022
892 Takip Edilen1.1K Takipçiler
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
I really take this for granted now, but in less than 30 mins, we: - generated a realtime collaborative app - deployed it into an authenticated & secure env - used the software together - and then let AI agents use it seamlessly on their own!
steve.prophet@nilslice

Fun times with @pjausovec playing his generated Tic Tac Toe game on Prophet... and then having our own agents play a round for us!

English
0
2
4
310
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
Fun times with @pjausovec playing his generated Tic Tac Toe game on Prophet... and then having our own agents play a round for us!
steve.prophet tweet mediasteve.prophet tweet media
English
0
4
2
900
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
so get smart about how you generate code... example token analysis of compilation output. model generates DSL / intermediate format, compiler turns it into actual runtime code. need to only generate 796 tokens to deploy 8,265 tokens. 10x+ token difference (!!)
steve.prophet tweet mediasteve.prophet tweet media
steve.prophet@nilslice

token efficiency is near-term super critical. benefits to cost & coherence are night & day when you can 10x efficiency. e.g. when generating code, you want the best deterministic output possible. the best outputs come from efficient token use.

English
1
3
6
1.4K
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
To my knowledge, I think we have the best abstraction and utilization of @cloudflare’s Durable Objects ever. They become an invisible layer to some seriously impressive primitives.
English
4
4
49
9.3K
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
Soon™️
English
1
2
4
609
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
things are getting weird
steve.prophet tweet media
English
1
1
1
445
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
We've entered a completely new era of app & system extensibility... and I'm more excited than ever about an AI-driven stack we've been building... more soon In the meantime, we are narrowing down our focus on what's new, and sunsetting the old. As such, mcp . run and XTP will go offline this week! Thanks to all who helped shape these products. I think you're going to really like what is coming next.
English
3
3
22
2K
dylibso retweetledi
dylibso
dylibso@dylibso·
Connect your apps to AI on your terms. Introducing Turbo MCP, our enterprise-friendly, self-hosted MCP security platform and connectivity gateway. If you're looking to bring AI productivity to your workplace, Turbo MCP was designed with partners in highly-regulated industries, ready to meet the most rigid security requirements. 👉turbomcp.ai
English
2
7
32
585.4K
dylibso retweetledi
AI Native Dev
AI Native Dev@ainativedev·
What if AI could extend your app’s capabilities? Steve Manuel (@nilslice), founder and CEO of @dylibso, joins @sjmaple on AI Native Dev to decode MCP and reveal how mcp.run lets developers safely scale AI beyond its original limits. He dives into how developers can extend AI beyond text, connect to diverse data sources, and design enterprise-grade workflows with confidence, all with MCP. Catch their full conversation now. Link in the comments.
English
2
6
5
1.2K
dylibso retweetledi
AI Native Dev
AI Native Dev@ainativedev·
The risk of letting AI do more than autocomplete? It can quickly spin out of control. On this episode of AI Native Dev, Steve Manuel (@nilslice), founder and CEO of @dylibso, unpacks MCP, the protocol that keeps AI extensions safe and predictable, and dives into mcp.run, his framework for tapping into shared MCP servers without losing control. With @sjmaple he shares: • why plugin-safe AI might be the most significant shift in developer tooling this decade • how mcp.run isolates compute to prevent AI-driven code from being unpredictable, letting developers experiment without risk • the practical and surprising ways web-connected agents expand what applications can do, from automating workflows to integrating new capabilities • why first-party code remains essential even in an AI-native world (00:00) Trailer (01:32) Deep Dive into MCP Architecture (04:50) Trust and Security in MCP Servers (21:11) Managed vs. Self-Hosted MCP Servers (26:48) Understanding Serverless and MCP Servers (29:10) Developing and Deploying on mcp.run (33:20) Security and Authentication in MCP (38:31) Future of MCP and AI Innovations
English
0
3
4
599
dylibso
dylibso@dylibso·
💪💪💪 Welcome, @shritesh!
Shritesh Bhattarai@shritesh

I've joined @dylibso as CTO. I've been a huge fan of the team for pushing software extensibility with WASM (Extism, ChicoryVM, XTP) and bringing the same vision to AI infra (mcp·run, TurboMCP). Pumped to help shape the next gen of AI products. Let's make software squishy.

English
0
0
2
437
dylibso retweetledi
steve.prophet
steve.prophet@nilslice·
Excited to be at #AIInfraConnect this week, showing off our new self-hosted, cloud-native enterprise MCP platform! It's not just devs, everyone is connecting everything to AI... if you want to streamline SSO + IdP integration, OAuth, DCR, and tool security, find our booth with this epic backdrop 👀
steve.prophet tweet media
English
1
4
10
1.9K
Robert Balicki (👀 @IsographLabs)
If you make it I'll instantly use it! Honestly this seems like something that will 10x the reliability of coding agents. And being able to go to definition will 10x their speed, if they're attempting to grep through massive files (eg our GraphQL schema is 10s of thousands of lines...)
English
1
0
0
33
Robert Balicki (👀 @IsographLabs)
Hey @dylibso, does mcp.run provide an MCP that provides a bridge to an arbitrary running language server? e.g. so that my agent can ask "what is the type of foo in `import foo from 'get_bar'` and the like?
English
1
0
0
144
Ben Patton
Ben Patton@benapatton·
@dylibso @kentcdodds @StatisticsFTW I think I am always missing something in this. Cause I agree those things are important but are we saying that UI's will disappear? Like businesses will not be started on the web anymore as we know it? I fail to see the 'interface' that businesses have with users
English
2
0
0
49
Ben Patton
Ben Patton@benapatton·
Having read this article now by @kentcdodds, I am wondering a couple things: 1. from our friends at @dylibso, I feel they are on track to do this with their setup 2. What is to prevent vectorization of tooling so that tools become context to be discovered and used? epicai.pro/mcp-search-eng…
English
1
0
0
112
dylibso
dylibso@dylibso·
@benapatton @kentcdodds @StatisticsFTW x.com/nilslice/statu… we still mostly think this — and it’s going to be inside ChatGPT, Claude, Gmail or Grok apps.
steve.prophet@nilslice

Thinking about what could happen to apps over the next stretch of AI-driven change… We should really just have a single “app”, and it connects to everything else. AI just-in-time generates a thin layer of UI inside this app, connected to APIs and tools. It’s probably going to be @x or @Meta or @gmail & all the AI labs will compete too — huge user bases, their existing apps already cover all the needed CRUD ops and related media + UI components. they have distribution and their own models to integrate, but must allow user to BYOM. So IMO it’s databases, tools & connectors, auth/identity, and composition of these that are the last really important layers of software. Really not that different than when we had all kinds of different single-purpose hardware which collapsed into the smart phone via apps: - handheld PDA - Walkman / iPod - cell phone / pager - digital camera - access badge - … The same has got to happen with collapse / consolidation of apps now… On-device AI will create another complete shift just like on-device apps did.

English
2
0
0
82