I see myself using less and less of Hermes agent from @NousResearch these days and more and more of Personal agents like @Muse & @hark_labs
Where did Hermes lag behind? Too much complexity?
Free AI models can lower the barrier to building a SaaS, but the model is only part of the cost.
Hosting, inference and distribution matter too.
Solo builders: where did you find the biggest unexpected cost?
I asked Hark from @hark_labs if I should buy some device like Omi or anything similar to really feel like in future or enhance Hark experience. It said below answer.
What's being cooked @adcock_brett ? I'm curious now.
Just moved my whole trip admin to @wajo_ai: flights, fare checks, even calling offices back home to sort things out. I just ask Fo and it gets done. Genuinely the most useful AI I use daily.
With LLMs getting faster, is multi-agent orchestration worth it for solo devs?
Most coding tasks are sequential and stateful anyway. Can't we just stick to single-agent workflows and simple prompt chaining, or am I missing out on major gains?
@itsjdraven honestly the craziest part isn't even the model itself, its that 1M context is casually becoming the standard free tier baseline. closed source labs are gonna have to drop their api pricing real fast to keep up tbh