News for hackers. Stuff that matters.

matteogrella

matteogrella on Hacker News

Show HN: Dwarfreflect – Extract Go function parameter names at runtime

Posted by matteogrella on Tuesday July 29, 2025 @09:39PM
from the show-hn dept.

https://github.com/matteo-grella/dwarfreflect

While working on openai-agents-go, I wanted users to define tools by passing in a plain Go function and have the agent figure out the inputs automatically.But I ran into a gap: Go's reflect gives you parameter types and positions, but not the actual names you wrote.So I built dwarfreflect: it parses the DWARF debug info embedded in Go binaries (unless stripped) to recover real function...

( Read More... | no comments | on HN )

Show HN: Go Port of OpenAI's Agents SDK

Posted by matteogrella on Sunday June 15, 2025 @04:57PM
from the show-hn dept.

https://github.com/nlpodyssey/openai-agents-go

Among the agentic AI frameworks I’ve tried, OpenAI's Python Agents SDK has the best balance of features and simplicity. Since I usually understand software by porting it to another language, I contributed to a Go reimplementation!It’s a work in progress, but already quite usable. Repo: https://github.com/nlpodyssey/openai-agents-goFeedback is welcome: on the design, missing features,...

( Read More... | no comments | on HN )
Re:Llm.c – LLM training in simple, pure C/CUDA on Tuesday April 09, 2024 @04:14AM (#39977181)
I'm the creator behind https://github.com/nlpodyssey/rwkv.f90. How about joining forces?
Re:RWKV.F90: Large Language Model in Fortran on Wednesday July 19, 2023 @04:22AM (#36783447)
Hi there,

Just so you know, I'm the one who wrote RWKV.f90. After making a lightweight autograd package that's a lot like PyTorch, but in Go (I called it spaGO), I am now passionately working on adapting the same concepts to the Fortran language.

But first things first, I want to make sure that using your GPU is as easy as pie. That's my next challenge.

Show HN: I've just ported the RWKV LLM to Fortran

Posted by matteogrella on Thursday July 13, 2023 @03:10PM
from the show-hn dept.

https://github.com/FortAI-Hub/rwkv.f90

RWKV.f90 is a port of the original RWKV-LM, an open-source large language model initially developed in Python, into Fortran. Known for its robust capabilities in scientific and engineering computations, the primary focus of this project is to explore the potential of Fortran within the realm of Artificial Intelligence.Please note that this is an ongoing project and I welcome contributions!

( Read More... | no comments | on HN )
Re:RWKV: Reinventing RNNs for the Transformer Era on Tuesday May 23, 2023 @07:55AM (#36043026)
As one of the authors, I'd like to clarify: the equations of the RWKV model enable computational parallelization, provided that the sequence is predetermined. This parallelization occurs during both the training and inference stages, specifically during the prompt reading process (consider it an "encoding"), right before the generation (or decoding phase).
Re:Show HN: A simple self-hostable Machine Translation service on Monday April 26, 2021 @01:51AM (#26939612)
Thank you! Should you have any question or curiosity, I'm happy to answer :)
Re:Show HN: Experiments on Machine Translation in Pure Go on Wednesday February 17, 2021 @03:47AM (#26164170)
I'm grateful to anyone who feels like commenting, critiquing, or even better getting involved to improve this Machine Translation implementation which is currently very slow (almost unusable except for very short sentences).