OpenAI is down! Good thing Rig agents let you easily switch to another provider like Anthropic, or more š
Or, you also can use the fantastic @ollama project to spin up llama3.2 API server. Local models will never let you down š
š§µThis is a big simplification of the Python ecosystem. I describe it more as a glue, making it easy to build and connect good code from all languages, connecting it to your favorite APIs and libraries.
10/10 | There are a ton of great reasons for choosing Python, Typescript, Rust, or any other software stack ā imo, the best language for the job is the language you know.
If you want to learn more about why I use Rust via Rig, drop a follow and look out for a future thread š.
9/10 | This is super valuable for developing LLM applications. Refactor away and as long as you appeal the compiler, all will be well.
With Rust, you can also deploy zero-cost abstractions, providing free DX for no loss of performance ā Rig leverages this heavily!
The #web3 data universe is expanding.
Thousands of datasets across multiple chains are already available on @graphprotocol.
Today, weāre excited to announce Async Subgrounds āthe utmost solution for faster data processing in pure #Python.
Learn more:
ā¶ļø playgrounds.network/blog/async-subā¦
As someone who has been querying protocol-level with Python for ages, seeing natural language surface these nested insights blows my mind š¤Æ.
It's incredible to see LLMs transform how we'll be interacting with the chain. It was a seriousš”moment when I actually ran this locally.
Premia Blue subgraph by @PremiaFinance is live on the @graphprotocol's decentralized network!
Subgraphs are open APIs, so anyone can access & query them, e.g. learn more about Premia Vaults.
With Playgrounds, you can do it through natural language!
The rig framework from the @arcdotfun team remains one of the most well-written ai agent codebases I've read so far.
Just fantastic work from the team to build out a solid core. Now it's time to expand on functionality.
8/10 | What language has elements of all these values? Of course, it's Rust!
Rust provides safety guarantees through it's strict compiler that leverages types, borrowing semantics, and errors as values that make building programs consistent and reliable once compiled.
2/10 | In general, agentic frameworks are wrappers on top basic web clients (requests, fetch, etc.) adding a set of abstractions that aid in managing state. This can happen in any language since there's nothing restrictive about the environment agents usually run in!
3/10 | Ofc, there are strengths and weaknesses in every language ā Python might not be the best choice for high performance, C++ might be heavy weight for simpler automations.
In the agentic world, most languages are good enough.