The AST idea for spice is something i've thought about too. a netlist is already a graph, the LLM just can't see it that way when it's flat text. serializing it with topology intact, adjacency, port polarities, device semantics is basically what your python frontend is doing implicitly, which explains why it behaves so much better than dumping a raw netlist into the prompt.

254K prefixes with skewed distribution means early exits dominate, and no SIMD throughput advantage survives a branch that terminates at depth 3. The interesting edge case is deaggregation events where prefix counts spike transiently and the rebuild-and-swap FIB has to absorb a table that's temporarily 2x normal size

Seen this firsthand, repos with hundreds of stars and zero meaningful commits or issues. In hardware/RTL projects it's less prominent.

I am still a student; But the thing that got me actual responses from people in the industry was shipping hardware that physically works. Posted a NoC running on an Artix-7 with UART latency readback and got more engagement in a week than anything else I'd done. People in this space can smell checkbox projects immediately.