Thanks! Yea it started as a bit of an anger management project. I am the absolute worst version of myself when forced to endure customer service calls.
1. The user (you) needs to provide it in the prompt for the call. I have ideas for being able to build memory context for each user so you wouldnt have to re-provide those details each time you want to make a call and instead pull it from memory bank, but thats not implemented yet.
2. So far I have actually never been called out for not being a human. I think thats in part because these people probably get a lot of pretty crazy calls and by comparison a slightly robotic over-the-phone voice has a lot of forgiveness. When you know, its easy to hear though. I did build it with Elevenlabs TTS to make the voice as natural as possible. Models have gotten much faster and so latency feels good, not outstanding
3. So this IS very interesting, I have to refine the system prompt extensively to get it to the right balance of objection handling without being weird and obviously non-human. It is very clearly on the side of the human, in fact in the NYT case it was initially told no but pressed ahead anyway. "I understand that, how do we move forward" "any other way we can resolve this" that kind of thing. Better still, customer service has a very high abr for hanging up the phone,so the persistency is very effective
Yes I think the API holds real potential. The space is busy, but I think there could be room for being the simplest possibel way to make a phone call programmatically. There would be less customization than a Vapi or Bland, but it would take 10 minutes to implement. Basically anywhere engineers experience automated workflows breaking because the real world requires a phone call. Voice probably wouldn't be core to these users business though, int hat case they'd want more control than we would offer.
Let me know if you have any issues trying it! Very curious to see how folks end up breaking it. We have a discord you can join here (https://discord.gg/2tn2ugXu) if you have any problems to troubleshoot (or email me!)
Yep, absolutely right! Combining our product with something like the carbon curing tech of Solidia or CarbonCure would make a net-negative concrete. I don't consider us competitors at all!
Impressive analysis! Missed out on a couple details, but I'll take that as a thoroughly researched/analyzed endorsement. Most importantly, we'll be scaling up a lot faster. Our current 1t/yr is just a demo. Our current roadmap is scaling up to a 300t/yr pilot in the next few months before setting up building a commercial plant (10kt+) in the second half of next year. Any chance we can poach you from Google? ;)
Absolutely! This is a big part of our thinking for how this integrates with future energy systems based on renewables. We'd offtake cheap electricity when oversupplied, like in California right now
Our process produces calcium carbonate, CaCO3, which cement producers can then burn to make quicklime, CaO. This does release CO2 but that CO2 has been captured from the atmosphere in the process of making that calcium carbonate. We only address process emissions (ie. splitting CO2 from CaCO3), but we do also produce hydrogen on-site, which can be used to replace fossil fuels in generating the requisite heat
Yeah absolutely, the CO2 balance comes from re-equilibration with the atmosphere. The rivers replenish calcium content on a longer time scale. But as you point out we're in no danger of running out of calcium in the oceans
The ocean is 20x supersaturated with calcium and is constantly being replenished by geological stores of the stuff coming in through rivers. As long as the ecosystem remains supersaturated this shouldn't be a problem for any marine life. Though a slight dislcaimer there is i'm not a marine scientist!
Absolutely, CO2 emissions from transport isn't much of an issue compared with the saving. Hopefully someone like Remora will crack that problem anyhow. We'll most likely need to build our plants closer to where the market is though, to limit the financial cost of transportation
Thanks! ABsolutely, looking at optimizing time of day we operate will be important. As grids become more dominated by renewables a solution like ours will be able to help stabilize acting as a electricity sink when the wind is blowing too hard or the sun too strong, taking advantage of lower power prices at these times
That's actually the trickiest part of the engineering, but we've managed to make it directly compatible with the an intermittent energy source like solar and wind
Love the enthusiasm to share! This is a bit of a coming out party for us, so not much out there in the way of public information. We'll add to that as it comes higher on the list of priorities
1. The user (you) needs to provide it in the prompt for the call. I have ideas for being able to build memory context for each user so you wouldnt have to re-provide those details each time you want to make a call and instead pull it from memory bank, but thats not implemented yet.
2. So far I have actually never been called out for not being a human. I think thats in part because these people probably get a lot of pretty crazy calls and by comparison a slightly robotic over-the-phone voice has a lot of forgiveness. When you know, its easy to hear though. I did build it with Elevenlabs TTS to make the voice as natural as possible. Models have gotten much faster and so latency feels good, not outstanding
3. So this IS very interesting, I have to refine the system prompt extensively to get it to the right balance of objection handling without being weird and obviously non-human. It is very clearly on the side of the human, in fact in the NYT case it was initially told no but pressed ahead anyway. "I understand that, how do we move forward" "any other way we can resolve this" that kind of thing. Better still, customer service has a very high abr for hanging up the phone,so the persistency is very effective
Yes I think the API holds real potential. The space is busy, but I think there could be room for being the simplest possibel way to make a phone call programmatically. There would be less customization than a Vapi or Bland, but it would take 10 minutes to implement. Basically anywhere engineers experience automated workflows breaking because the real world requires a phone call. Voice probably wouldn't be core to these users business though, int hat case they'd want more control than we would offer.
Let me know if you have any issues trying it! Very curious to see how folks end up breaking it. We have a discord you can join here (https://discord.gg/2tn2ugXu) if you have any problems to troubleshoot (or email me!)