etgibbs

15 ms·

1.
by etgibbs 1y ago
good suggestion, adding to the next release
2.
by etgibbs 1y ago
awesome!!
3.
by etgibbs 1y ago
we are working with this consumer electronics company debug their zephyr project right now, would be worth trying out. if it doesn't work, email founders@embedder.dev we'll help you out
4.
by etgibbs 1y ago
you can upload the esp32 docs to your own project, but we are working with espressif to have first party support for the docs baked in
5.
by etgibbs 1y ago
exposing the context layer as an MCP is a very interesting idea that I did not consider; could be a good way to drive adoption. great suggestion.
6.
by etgibbs 1y ago
interesting, would you be open to chatting?
7.
by etgibbs 1y ago
post cannot be edited, unfortunately... thanks for the dev kit :)
8.
by etgibbs 1y ago
we are talking to a some of the larger vendors about getting first-party support for documentation. once we scale up a bit I think it could make sense to have multiple fine tuned models based on the docs they provide (or we crawl the web fo
9.
by etgibbs 1y ago
there are some other companies trying to solve for PCB design. long term we want to build end-to-end AI automation for those cross-domain projects. this was a good place to start based on our personal/work experience
10.
by etgibbs 1y ago
need to update the policy, this was for the original version of the app, which used a deterministic algorithm to generate driver code, which we stored on cloud so you could download the files. thanks for pointing this out.
11.
by etgibbs 1y ago
we store the files you upload for indexing we don't store generated code / data, but it does pass through our API to the model provider. we store the usage meta data for billing purposes
12.
by etgibbs 1y ago
strict grounding protocol + planning phase, mostly by prompting and forcing attention through citations. it tends to think longer than other coding agents but the results are usually better. let me know what you think.
13.
by etgibbs 1y ago
great question. we've found most code/docs are offline in the embedded space (for good reason) so our approach going forward is going to be more FDE/on-prem for enterprise users. they asked for self-hosted, BYOK, local indexi
14.
by etgibbs 1y ago
currently we scaffold frontier models. the product is basically a context layer with custom tools that enable hardware interaction. we've tossed around the idea of pre-training/fine tuning but new models are being released so fast