1,478 karma · joined March 25, 2015
I was interested in building a product for which 1. the agent is the whole product, not just a component of it; and 2. solves a specific problem out of the box.
Tavi is a deep people search agent that lives in Slack. We used it to find our founding engineer and our first customers are a mix startups, recruiters and VCs.
While we are focused specifically on technical roles in AI startups, and we don't "verify" per se, we do enrich jobs data with investor-grade intelligence on the startups themselves, so you can see which companies have legitimate backing.
The project has been a huge learning curve for me - I started out as a skeptic of how generative AI could solve real problems (rather than just create noise) but now think that, like the internet, it can create a new kind of abundance that will be harnessable in all sorts of interesting ways.
With regard to job posting accuracy - absolutely this is a problem. Normalising what different job descriptions actually _mean_ is one of the key challenges we want to solve.
Which additional pieces of job information do you think we should surface before clicking?
Agent is definitely an area I'm looking to improve with persistence and make it truly personalised.
Real estate and crypto on the other hand...
If you think wanting to live in a unipolar world is "evil", then there's no helping you.
Every technological development since the industrial revolution has increased human demand for energy in some way. It's only the environmental movement, which actively shut down nuclear power plants, which wants human energy consumption to be reduced for cultish reasons.
If we'd ignored anti-nuclear activists in the 70s none of this would be a problem.
I have some experience with accounting software and its use in various types of business (not just startups).
Xero (rather than Quickbooks) is dominant in the UK and it has a lot of problems. However, I think you are addressing a fundamentally different market. Quickbooks and Xero are both targeted at the long-tail of small businesses - think tradesmen, cafes, hairdressers, etc. The main thing they accomplish for those customers is basic book keeping and business functions (like invoicing, payroll, etc)
Your software seems much more aimed at small companies which want sophisticated tools to generate financial insight. This is not a criticism, just a comment that this is a different and potentially smaller market.
However, one way you could capture the "long tail" of small businesses is by winning over the high-street accountant. If you could let them prepare annual accounts with 10x less work - even if the customer book keeping is a mess - then every accountant will recommend your software to their clients.
Anyway - great to see something new in this space. Feel free to reach out if you'd like to discuss further.