I wonder if you would need to expose each command as its own tool. Might be able to expose a single tool with a description and context on how to discover and call commands therein?
26.05.2025 07:45 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0@russcam.bsky.social
Software Engineer with keen interest in Search/Information Retrieval, Scalable Distributed Systems, Observability & Performance, and Software Engineering Leadership. Also BMX/MTB, Climbing, Photography, Guitars!
I wonder if you would need to expose each command as its own tool. Might be able to expose a single tool with a description and context on how to discover and call commands therein?
26.05.2025 07:45 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0For how to work with command line tools built with System.CommandLine e.g. built-in commands, conventions for args and options?
26.05.2025 07:16 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Some fun building an MCP server to expose tools for querying any search system, judge candidate results and calculate relevancy metrics like NDCG and ERR:
www.youtube.com/watch?v=vLhp...
It inferred tools to call and how to build queries
#search #relevancy #evaluation #metrics #mcp
May want to take a look at Brave and Arc
23.05.2025 14:02 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Totally! The key is framing improvements in terms of business value- revenue impact, conversion rates, retention. "Better search relevance" isnโt compelling; "X% increase in purchases due to Y% improved relevance" is, with experimental results to back it. Make it about outcomes.
07.03.2025 02:49 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0๐ Introducing Search Pioneer Ranking CLI โ An open source Learning to Rank (LTR) command line tool for training rankers with LightGBM and FastTree:
github.com/searchpionee...
โ
 Train rankers with LightGBM & FastTree
โ
 Calculate feature importance using PFI
โ
 Split and Fold data sets for training