Victor Grdr

@samrith such a great idea! upvoted:) super curious about how are you handling the actual mcp protocol implementation? we've had so many issues with mcps being flaky or not maintaining connections properly. also curious about latency - are you caching query results or is every agent call hitting the db directly? and last thing, how do you handle when different agents need different levels of...

i love going down wikipedia rabbit holes but yeah the experience could be so much better. curious how you're pulling in the wikipedia data - is it through their api or are you doing something custom? also wondering if claude can cite specific wikipedia articles in the responses or if it's more of a general knowledge thing. we're working on connecting our agents to external knowledge bases and...

okay this is exactly what i've been complaining about for months - why do i have to explain the same context to every single llm? love that you're syncing chrome and twitter bookmarks too, that's smart. question though: how are you handling the memory sync technically? is it through their apis or are you doing something more clever with mcps? i've had so many issues getting context to persist...

this is super cool, i've been dealing with mcp hell lately for my own startup and honestly the setup process is brutal. curious how you guys are handling the crawling and indexing - are you doing real-time updates when sites change or is it more of a snapshot thing? also wondering about rate limits and how you deal with sites that don't want to be crawled. we've been trying to connect our...

@kirillgreen congrats on the launch, super curious if you have thought on providing personalized summaries so it fits better the readers' preferences?

Congrats on the launch @stephanschulz ! Love the "recommended based on interests" piece - tailored learning paths only work if the app actually knows the user. Curious how you're tracking what someone's interested in vs just what they clicked on. YC founder here, researching user context for AI apps.

Congrats on the launch @tommy_bar_av1 ! Action-taker > note-taker is spot on, actually extracting what matters is pretty daunting task (i'm doing memory for agents) curious how do you decide what's signal vs noise? onwards!

congrats on the launch @arvinx love the "agents that grow with you" framing. The persistent memory piece is exactly what's missing from most agent tools today/ curious how you're handling state persistence across conversations? Building something adjacent at mio.xyz. Congrats on the launch!