
Senior Staff Engineer @thekonginc | Open-Source 🌱 DX & UX 🛟 Travel 🛫 You can also find me on 🦋
Louisville, KY
Joined March 2010
- Tweets8.2K
- Following1.1K
- Followers457
- Likes1.4K
Adam DeHaven retweeted
Replying to @willblaze
Imagine what it feels like never having to log on LinkedIn again
Adam DeHaven retweeted
At Kong we run half a quadrillion AI tokens a year, and I kept seeing agents stuck on clouds built for websites.
So we built a cloud for agents.
Volcano is live today: the self-healing agent cloud. Sandboxes, durable workflows, Postgres, hosting.
No one wants to use your agent
We want to use your thing with our agent
DUFFY: How many times does this have to happen before congress gives the money to fix it?
R: You’ve been in office almost two years now, Republicans controlled both chambers of congress and still do.
DUFFY: What your network are you from?
Adam DeHaven retweeted
Replying to @RaillyHugo @ctatedev
huge thanks to @lahfir for the video and @BenSabic for the changelog + chat sdk side 😁
vercel.com/changelog/json-re…
This is the most powerful and fastest Comark we've ever released 😉
Thanks to all the contributors!
:: Comark 0.9 is out!
• Change any node at render time, in every renderer
• A smaller parser with no default plugins
• New heading-links plugin
15+ fixes across parsing, Markdown round-trips, and renderers.
comark.dev
Adam DeHaven retweeted
HUGE list of AI features announced at Summit yesterday:
▪ AI Cost Management
Visibility, attribution, and control to make AI spend defensible to finance and prove ROI (plus monetization capabilities)
▪ Context Management
Create and manage Agent access to context via cost, performance, and quality-optimized MCP server generation on top of APIs, other MCP servers, and enterprise data stores.
▪ Token Vault (Agent Identity)
Kong becomes the holder and the broker of agent credentials. Agents get access to every system they need without ever possessing a credential.
▪ Registry & Catalog
The Registry is where agents discover and subscribe to what they're sanctioned to use. The Catalog is where humans keep governed inventory of everything running. Together they are Kong's AI-native discovery surface.
▪ Advanced AI Observability
Trace every agent conversation end to end — every turn, tool call, and policy decision, with cost and latency at each hop.
▪ Webhook Engine
Build and govern agent reactivity.
Congrats to Marco Palladino , Reza Shafii , Saju Pillai, and the entire product and eng teams for creating the industry's best AI control tower. Onwards!
The new Portal MCP Server from @kong makes your Dev Portal content available through MCP, giving agents a structured way to discover APIs, read documentation, and understand how to use your services.
Excited to see this ship for both developers and agent konghq.com/blog/product-rele…
2
3
147
Adam DeHaven retweeted
Markdown as a templating engine ✔️
Coming soon in the binding plugin for @comarkdown
The new Konnect Dev Portal Toolkit from @kong brings live previews of your Konnect Developer Portal right into VS Code✨
Pair it with the Konnect MCP and let AI agents author, brand, validate, and visually verify Dev Portal content on their own.
konghq.com/blog/engineering/…
when i used to learn a new library or framework i had this urge to come up with an abstraction the moment i hit something i didn't like
this was always wrong. i should have turned off my brain and just focused on absorbing the different way of doing things
when you do that you realize most of it makes sense once you've adjusted
the same thing is happening right now with ai coding. everyone is coming up with some incremental trick or patch for these tools perceived short comings
and they're very proud of them
but now is the time to absorb not fix
We're introducing Claude for Teachers: free access to premium Claude capabilities for verified K-12 educators in the US, with a library of teaching skills and a direct connection to evidence-based curricula, mapped to academic standards in all 50 states.
claude.com/solutions/teacher…
Interesting take on gamifying transparency on Slack
We're killing the DM at @Zapier. Starting with the executive team.
We've long held Default to Transparency as a value. That value has largely encouraged communication in public channels. But as the company grew, DMs are a hard habit to resist and break.
But every DM is a gap in our Shared Brain. It's context that is lost for humans and AIs. As a result the cost of DMs keeps going up.
So earlier this year I posted about our exec transparency leaderboard. The leaderboard has become quite the competition internally…
I'm 3rd today. My co-founder @BryanHelmig has held the top spot as long as I can remember…
It sets a standard for the rest of the company. In fact, since last year we’ve seen the % of Slack messages in public channels go from 33% to 46%.
What the leaderboard measures
Transparency is a team sport, and a disinfectant. Every month we track what percentage of our execs' Slack messages happen in public channels versus private DMs.
When your CEO debates strategy in a DM, that decision is invisible to every agent and every team that needs to know what was decided and why. The decision happens but the reasoning vanishes.
When that conversation happens in a channel, it stays. New hires can search it, agents can read and verify it, etc. Your Shared Brain knows what's true now: ask it a question and the answer reflects the latest reality.
Taking It to the Next Level
Reducing DMs are one way to increase transparency and open up context for humans and AI, but there are other mechanisms that help too. Three things beyond the leaderboard:
1. Meetings get recorded, transcribed, and become queryable
2. We run a shared skills library. Anyone on the team can encode a workflow they've figured out into a skill and share with the team
3. And we keep score. It's a silly scoreboard, but it subtly drives positive behaviors
Raising Your Ambition
In order to get the most of AI in your company, the AIs need context. So making your context queryable is one of the most practical moves you can make to improve the effectiveness of your AI agents.
P.S. I’m coming for #1, Bryan...
103
Adam DeHaven retweeted
If you’re deploying to prod 4500 times a day, maybe one of those deploys could fix the bug where my car shows the name of the previous song instead of the current one when playing from Spotify over Bluetooth? 😅
Am I the only one who uses @Spotify ~6-8 hours a day and sees zero change? Was that disco ball icon change 4,499 deploys?
Boris sat down with Spotify VP of Engineering Niklas Gustavsson.
Spotify ships 4,500 production deploys a day, and 73% of PRs are now AI-assisted.
2
3
531























