Reposeek

Aurelie Purdy 1 month ago

Honestly struggling to see the market fit here. GitHub's built-in search isn't great, sure, but is it bad enough that developers will switch to a separate tool? The real question is what happens when GitHub inevitably improves their own search ranking algorithm. That's the existential risk. Also curious about monetization, is this heading toward a freemium model with paid features, or purely ad-supported? The AI angle sounds nice in theory but "analyzes various factors" is pretty vague. Star count and fork count aren't exactly groundbreaking metrics to consider. 🤷‍♂️ Would like to understand what the AI is actually doing beyond weighting existing public data.

Myrna Batz 1 month ago

Huge congrats on shipping this! GitHub's native search is honestly painful when you're looking for something specific and end up sifting through abandoned repos with 5 stars. The quality ranking is what really sets this apart, star count alone tells you nothing about whether a project is still maintained and worth investing time in.

Gregory Emard 1 month ago

Curious about the ranking model, are you using a weighted composite score or is there an ML layer trained on engagement signals? Would be interesting to see how you handle repos with low stars but high relevance to niche queries 🤔

Linwood Feeney 1 month ago

Clean typography and the quality-weighted ranking is genuinely useful, but the results page needs more breathing room between repo cards.

Murl Block 1 month ago

stars and forks just measure popularity. what underlying model actually determines quality?

Loren Halvorson 1 month ago

why use this over github's own search

Stephany Ratke 1 month ago

what embedding model are you running for the relevance scoring?

Brook Feest 1 month ago

githubs search is so bad this actually matters

Rosalyn Bergnaum 1 month ago

clean spacing makes browsing repos way less overwhelming

Luna Sporer 1 month ago

just github search with ai sprinkled on top. revolutionary.