Hasura now in GA
supergraph: a query planner across all your databases
metadata: declaratively configure your domain
for teams: independent SDLC for subgraph modules
AI ready: user level permissions at the metadata layer
API: graphql, and more...
@HasuraHQ
An AI agent can only be as smart as the data endpoints it's connected to. That's why smart people connect smart agents directly to the database.
Authz? Define permissions directly on your data. Now ALL queries are safe.
The way we build our backends is SO not ready for the future.
AI agents need direct access to the domain models and the core business methods.
The way we've been building APIs is for "dumb clients", with API endpoints that each contain a lot of business logic.
Dumb clients need smart endpoints need dumb pipes to data. (ref: microservices).
Smart clients need no endpoints need smart pipes to data.
20 years of the way we build our backends, and the investments we've made, will have to completely change.
It's like if invested $10M in upgrading all your office infrastructure to use wired ethernet in 2005 and then iphone/ipad dropped in the next 2 years and you were like....oh shit.
LLM's can't have direct database access because: what about data security.
Declaratively define user-level permissions at the supergraph level. Suddenly, the LLM is the USER's agent.
Unlock AI powers with a consistent, standard, and expressive SQL API across your supergraph.
This works today. A live demo by @tanmaigo at @aiDotEngineer
https://t.co/KqrsjninbN
It's here. Run a complete supergraph with Hasura DDN on your own machine.
It's just docker containers and http servers. Self deploy with no limits. Open source components serve your API.
https://t.co/MT2lF1rBv6
The supergraph models your data domain. Tell it about the shape of your data and get your API for free.
The Hasura engine plans and executes optimal queries over disparate data sources. Connectors bring your own sources to the supergraph.
https://t.co/4SLDCMFcDw
Day 2 of hacking on developer tooling for @HasuraHQ
Today I'll be delving into iteration workflows with command line tooling #buildinpublic
https://t.co/ockUY5441g
Join me as I build and discuss developer tooling for Hasura v3 DDN. Write some code, review some spec, see some workflows. #buildinpublic@HasuraHQ
https://t.co/sCzHDh5zaS
📢Announcing the 9th edition of GraphQL Bangalore meetup
🗓️ January 30, Tuesday 📷 6PM to 8PM IST
📍 The Terrace at @HasuraHQ
We have talks from @mediocreDevops and @_nullxone.
RSVP here - https://t.co/L3kx9gxPdk
Besides the fact of its popularity, did using the TypeScript compiler have any other benefits?
I've been thinking that it should be possible to encode many constraints in the type system. Does the system generate any types that describe the real current state of the system?
Looks incredible :) like a declarative mash up of Terraform and Migration tooling. And it builds forms?
Besides the fact of its popularity, did using the TypeScript compiler have any other benefits?
I've been thinking that it should be possible to encode many constraints in the type system. Does the system generate any types that describe the real current state of the system?
Looks incredible :) like a declarative mash up of Terraform and Migration tooling. And it builds forms?
How do you grow communities?
- Make sub-communities, empower and recognize members
- Don't be afraid of losing control of the brand experience
- Use data for member retention
https://t.co/xibyiy4x1N