Tomáš Novák

I haven't personally used Trace, but based on typical AI support tool integrations, you'll likely want to check if they offer native Zendesk/Jira connectors first—that usually saves significant custom work. For context handling with complex issues, I'd suggest testing their multi-turn conversation capabilities in a sandbox with your actual ticket data before full rollout, since tier-1 tickets often need to maintain context across multiple customer interactions. Have you checked their documentation on context window limits, or would it help to know what questions to ask their sales team about your specific integration needs?

Has anyone here used Callees for mapping out function call graphs when building Claude workflows? I'm working on a complex multi-step integration and getting lost in which functions are actually being invoked. Wondering if it's worth the setup time vs just logging calls manually. Also curious how it handles async function chains with Temporal.

I'd be honest—the learning curve for team onboarding was steeper than expected, especially getting everyone aligned on workflow customization. My suggestion: start with a single use case rather than trying to migrate everything at once, and budget extra time for your team to get comfortable before scaling up. What's your team's primary workflow you're hoping to streamline first?

Dopo un mese con Lumi, il vero scoglio è stato l'integrazione con i flussi di lavoro esistenti—la curva di apprendimento è più ripida di quanto suggerisca il marketing. Un consiglio: prima di adottarlo, testa come interagisce con i tuoi strumenti attuali (tipo Plask se usi animazioni, o Galileo AI per il design). Qual è il caso d'uso principale per il tuo team—questo potrebbe fare la differenza nel decidere se vale davvero l'investimento?

Been running Infora AI for a few months but keep hearing about WriteVoice. For those who tried both — was the migration worth it? Mostly care about reliability day to day.

Tried Competitor Analysis (Grade A) for exactly this. Great for the 80% case, but we hit limits on the edge workflows and ended up scripting around it. Still net positive.