OpsCost was an earlier project of ours, and we're still running and maintaining it, though new feature work is obviously put on hold for the moment. We're glad you like it!

Without being involved in the process, it's hard to speak as to why healthcare.gov has had the trouble it has. But what's definitely true is that healthcare.gov was taking on a monumental job; they have to check income, eligibility, immigration status, and a hundred million other tiny provisions of the ACA, and the ways the ACA interacts with existing laws. And they're working with legacy databases, that may or may not have well-developed APIs, bugs, performance issues, etc. It's huge. And they had to operate at massive scale (it's a lot easier to scale up if you can do it gradually, as your traffic increases; having 250,000 users on day one is a nightmare, and it was five times what they expected). HealthSherpa's scaling up fast, and steadily, but we have time to measure our traffic, plan, and adjust. We're very consciously working on the smallest slice of the problem we can - the comparison-shopping experience, which is what's most important and most accessible to consumers.

We're obviously exploring options for new features, etc., but we don't currently have any plans to offer signups directly. (We've talked about it - we've talked about a lot of things - but it would be a lot of work to build). And there's one very important way that third parties can't step in for the official exchanges - people who qualify for subsidies must go through the official exchanges in order to get them (because only the official exchanges can do all of the verification work necessary to provide the subsidy). We're excited to see how healthcare.gov looks after the current round of development efforts, because for low-income customers, it's absolutely essential that it work and work well.

Yeah, we made a special project to pull in California data, partly as a proof-of-concept for how to get the other 15 missing states, partly just because California's got such a huge population that we felt it was important to have ASAP. We're working on getting all the rest.

Source: I'm on the team.

We're also not as young as all the articles say :). The articles saying we're all 20-year-olds is a misquote of the original CBS piece that called us "twenty-somethings". We're all in our mid-to-late twenties, and we've all previously worked for at least one big company you've heard of.

I'm pretty sure the pre-existing condition rules don't cover demonic possession, so you might be out of luck on that front.

We're working on getting more plan details. It's not included in the same dataset, so we have to go look for other sources for it. We've already got several we're working with, so as soon as we can get them integrated we'll have that info up.

Source: I'm a member of the team.

Working on it. The details aren't in the main data set, so we've got to go get those elsewhere. We've identified a few sources and are working on integrating them.

Source: I'm a member of the team.

No, it really is thanks to Healthcare.gov. The open access to their data is what made it feasible to build HealthSherpa - getting that data otherwise would have been an absolute nightmare. You're right that there are a few pre-existing sites to help people buy insurance, but even those mostly aren't offering ACA plans - and it's a lot harder to estimate premiums on non-ACA stuff.

Source: I'm a member of the team.

You're totally right that a 100,000 users would be a problem. But we've got a lot more than one server (we would've been slashdotted otherwise :p). (For comparison: healthcare.gov anticipated 50k at launch, got 250k, but their login service tipped over at 1100). We've got background from Twitter, Pinterest, and some very large e-commerce sites. We've got the infrastructure in place right now to scale up 10-100x beyond our current levels, and the know-how to go past that.

Source: I'm a member of the team.

What were the zip codes? I'll look into it. (I'm a member of the team).