On to a new ai project, this time an ai research assistant for financial analysts.
We have some built in Perplexity. I am building on top of it.
What we are building sits next to Bloomberg and AlphaSense in the market, if we can get Bloomberg to work, it depends on the licence that we hold. Proprietary data, AI doing the analysis, answers that used to take a junior three days come back in three minutes. I am not telling you the industry. I am not telling you the use case. The firm has spent years on this and I am not handing competitors the playbook in a Tuesday blog post.
The customer pays for an answer, not a tool. They want a question answered that they are already paying people to answer. Faster. Cheaper. With sources they can cite when their boss asks them where it came from.
I am committing publicly: paying customer in 90 days. Not a demo. A real version a real customer writes a real cheque for. Build-in-public means I will post the milestones, the misses, the deals that close, the deals that do not. I will not post the recipe.
Find the person who is 80 percent of the way there, add the 20 percent that makes it a product someone pays for. This is the 80. I am the 20. The same play I have been running on the AI employee build and on every fractional AI ops engagement that has come through TestVentures.
If you are a founder sitting on something built but stuck before it becomes a product people pay for — that is the conversation I want.
---
Monthly Revenues $11,800 | Clients 2 | Prospects (AI employee live this week) | Team: Me + Jan (CTO)