swapnil chand retweeted
Learning pipecat's internals by asking ai then yapping to myself and even taking handwritten notes.
making your own software is tough, even with AI, because it's hard to know what you really want
I didn't end up sticking with this, the modes weren't the right way for me to organize my work
gonna a different take that’s more to-do oriented
aigrants.in/?ref=MK9JXEHN
Use this to get free API keys via @aigrantsindia!
31
swapnil chand retweeted
.@aigrantsindia is looking for a deserving recipient of $5000 GPU credits from @jarvislabsai from @e2enetworks Tag someone you know who needs this!
cc @0xBosky @vaibhavbetter
had ai test every feature of the video kit for kencomputer.dev today. dubbing, captions, cleanup, scripts in your style. 34 of 34 passing now
best catch was iphone hdr clips coming out grey lol. fixed
1
87
Lol this is hilarious but so frustrating. Had something similar to me when I asked my agent to upload on insta it would just play a video and shocked randomly
Last night I was awoken by my agent. I asked it to reverse engineer airplay over night it decided at 3 AM that the right way to verify that it works was to listen on the laptop mic to a sound played at max volume in the living room. It raised the volume to hear it.
55
swapnil chand retweeted
What are the "missing benchmarks" of AI?
I'm pretty interested in the areas where, after years of advancements, AI still totally fails hard. Below is one -- AI in a group of people collaborating. Another is original comedy (think Seinfeld original genius comedy, not rehashed dad jokes or shock value humor).
The first really big adoption of conversational voice AI was in simple, customer-facing interactions: customer support, scheduling, and semi-scripted outbound calls to do things like remind people about appointments or qualify sales leads.
There's now a lot of interesting new stuff being built that pushes into more complex use cases and architectures.
Voice experiences have to: 1) be fast, responsive, and natural, 2) perform tool calls and instruction following very accurately across long, multi-turn interactions.
Most of the voice applications we build, today, use multiple models doing different kinds of inference in parallel, to achieve both 1 and 2. Here's a demo from Jon and Dom showing multiple people talking, with a voice agent in the call weighing in only when it's spoken to directly, or there's a semantic gap in the conversation. This uses Jev alongside a couple of LLMs.
swapnil chand retweeted
Anybody in my network who's connected to @sama? Need help, cold mails have not worked so far.
Done for the day. Damn. My bot will be creating some reels and tweets based on what I did today and share.
Instagram.com/swapnilbuildss…
1
132
Havent started watching gta 6 streams online yet. Will let it be. Enjoying the joy of going viral and using the second wind to build and then crash
4
205
Adding features to kencomputer.dev so that it can do dubbing and cloning and more for your voices.
Thanks to @ElevenLabs
2
195
swapnil chand retweeted
Can anyone help me build multiple TikTok and Instagram accounts from international numbers?
/urgent

















