rockwotj

If it’s like vitess then no, but IIRC you get relaxed consistency across shards

To me, this is like asking if having built a sql engine if you write SQL differently.

I think the answer is yes. Having a strong understanding of the underlying engine helps you debug and optimize more quickly

Needing reference stuff never changes unless it’s so frequently used.

I did this in about 400 or 500 lines of typescript with direct API calls into vertex AI (using a library for auth still). Supports zod for structured outputs (gemini 2.5 supports json schema proper, not just the openapi schemas the previous models did), and optionally providing tools or not. Includes a nice agent loop that integrates well with it and your tools get auto deserialized and strongly typed args (type inference in ts these days is so good). Probably could had been less if I had used googles genai lib and anthropic’s sdk - I didn’t use them because it really wasn’t much code and I wanted to inject auditing at the lowest level and know the library wasn’t changing anything.

If you really want a library, python has litellm, and typescript has vercel’s AI library. I am sure there are many others, and in other languages too.

How is this better?

Looks similar to absl::NoDestructor

https://github.com/abseil/abseil-cpp/blob/master/absl/base/n...

Which is basically the only usage of std::launder I have seen

Investors are investing just as much in the person/team as what they are building. What they build will change, but having a good team is all the difference.

It’s because vertex is the “entrrprise” offering that is hippa compliant, etc. That is why vertex only has explicit prompt caching and not implicit, etc. Vertex usage is never used for training or model feedback, but the gemini API does. Basically the Gemini API is Google’s way of being able to move faster like openai and the other foundational model providers, but still having an enterprise offering. Go check Anthropic’s documentation, they even say if you have enterprise or regulatory needs go use bedrock or vertex.

Can you elaborate? Are you conflating node and javascript?

Just write the for loop to react to tool calls? It’s not very much code.

Its pretty heavy in Azure, so I would be surprised if it went away. This is DBX play to move into the transactional database space in addition to the analytical database.

neovim

Copilot is limited to 64k context window. Even if the underlying model is gemini with 20x that. It’s gotta be a major reason copilot is so bad in comparison. They are all the same sets of models under the hood

The open source repo to generate an MCP server from a gRPC service is here: https://github.com/redpanda-data/protoc-gen-go-mcp/

Would be cool to also have a tool that uses gRPC reflection, then there wouldn't be any need for codegen!

Its a rust port of Redpanda Connect (benthos), but with less connectors

https://github.com/redpanda-data/connect

Yeah I have read gemini pro 2.5 is a much bigger model.

Part of it is probably tgat claude is just better at coding than what openai has available. I am considering trying to hack in support for gemini into codex and play around with it.

yes

My thoughts is this model release is driven by the agentic app push if this year. Since to my knowledge all the big agentic apps (cursor, bolt, shortwave) that I know of use claude 3.7 because it’s so much better at instruction following and tool calling than GPT 4o so this model feels like GPT 4o (or distilled 4.5?) with some post training focusing on what these agentic workloads need most

Looks exactly like ollama but built into Docker desktop? Anyone know of any differences?

And OpenAI doesn't/wouldn't if they had the chance?

It’s not really copy and patch, the whole point of the copy patch is so you can inline that in your compilation output and it’s a fast baseline interpreter because individual builtin functions are optimized (via the compiler output you’re copying from) and inlined (which is why you need to patch to update what registers are being used. In this model you jit only control flow really, then inline the implementation of each bytecode operation (in contrast to sparkplug [https://v8.dev/blog/sparkplug] which just calls a builtin instead of copy/patch). It’s still JIT which is vastly different than an interpreter.

> JIT will be things of the past

Sorry no. JIT is not going anywhere. They mentioned in the article JIT would be better performance just more effort than they wanted to put in (a good tradeoff!) JIT powers Java, Wasm and Javascript VMs and are certainly the way to get the fastest code because you can give the CPU code that it can do a much better job predicting the next instruction. With interpreters you’re often limited by the indirection of loads when looking up the next function to call, and generating code for the control flow outside calling your “builtins” is precisely what Sparkplug is doing.

At the end of the day, like most of engineering, choose the right tool for the job, which in this case is simplicity (which is often the right choice!), but that doesn’t mean it’s always the right choice. For example if browsers did this then Javascript performance would tank compared to what we get today.

The resulting VM implementation reminds me of this post from cloudflare where they similarly use closures to build their interpreter.

https://blog.cloudflare.com/building-fast-interpreters-in-ru...

I am hoping MCP will fix this. I am building an MCP integration with kapa.ai for my company to help devs here. I guess this doesn’t work if you don’t add in the tool

> Anytime Meta, Oracle, Microsoft, or anyone else announces another AI bit barn, a juicy power purchase agreement usually follows. Meta's mega DC being birthed in the bayou was announced alongside a 2.2GW gas generator plant — so much for those sustainability and carbon neutrality pledges.

Ouch, I grew up where Meta created its first data center and while it is very efficient (https://engineering.fb.com/2011/04/14/core-infra/designing-a...) I remember a more recent article about how they are zero net emissions (https://tech.facebook.com/engineering/2021/11/10-years-world...). I wonder if they have kept that in this AI rush (along with most of the datacenter companies).

EDIT: a direct link to net zero for Meta: https://tech.facebook.com/ideas/2020/9/facebooks-path-to-net...

I do like this approach in theory, but agreed I don't think that the devex has been solved and I don't know how many people would value a different approach as SQL feels like king in many places... FWIW it's kind of the approach that IndexedDB in the web takes, however that the API is quite bad IMO.

Or maybe there is a higher level DSL that you could apply to create query plans (something like MongoDB aggregation pipelines maybe?), but it quickly becomes basically the same as SQL.

is anyone outside of the research labs fine tuning models for production use cases? I have been seeing more people just using foundational models off the shelf especially in light of a new advancement that seems to come every few months

Redpanda Connect is a yaml file that has a cdc input and you can process the data and send it anywhere, no Kafka required.

I’ve seen postgres cdc directly being written to Snowflake.

https://docs.redpanda.com/redpanda-connect/components/inputs...

Why confluent instead of something like MSK, Redpanda or one of the new leaderless, direct to S3 Kafka implementations?

Maybe they are conflating the Starcraft success that Deepmind had with AlphaStar?

Generally if you care about safety then “pretty safe” doesn’t cut it.