Announcements
Introducing Imora: Build Your AI Twin
What is Imora?
Imora is for people whose judgement is the product and whose time is the limit: consultants, fractional executives, advisers, founders of agencies and boutiques, and founder-CEOs of small firms.
You build a twin. The twin answers your team and your clients the way you would decide, and it shows why. Every answer opens with the twin's read of the situation, and anyone can open "why this answer" to see the principles and evidence it used and how confident it could be.
The language models underneath are Anthropic's Claude. What is Imora's own is your reasoning graph and how it retrieves your decisions.
Why we built this
If you sell judgement, you are the bottleneck. A client wants to know whether to push back on a supplier's price rise. A junior wants to know whether a request counts as scope creep before replying. A new hire wants to know how you would handle a candidate who interviews well but has a patchy reference.
You have answered most of these before in some form. They still wait for you, because nobody else can say how you would call it.
A custom GPT with your documents and a persona prompt gets part of the way there, and for many jobs it is enough. It can repeat what you have written. It cannot show how you decide, and a correction lasts until the next conversation. We wanted something people could check.
How it works
- Answer five real scenarios. About 10 minutes. You say what you would do and why, in your own words.
- See what Imora took from them. It distils your answers into principles, each with the conditions it applies under, and shows your own words beside them.
- Chat with your twin. Ask it something new and open "why this answer".
- Correct it. Mark a reply "Not what I'd say" and give your version. That correction outranks everything else. "Sounds like me" confirms a good one.
- Share one link. Send it to your team or clients. On Pro you can also embed it on your site.
A longer guided interview (six sections, 30 to 40 minutes) and your own documents add depth. Both are optional. The full walkthrough is on the product page.
The reasoning graph
Your answers, corrections and documents are stored as evidence. Imora distils that evidence into conditional principles, such as "discount for a longer commitment, never for a louder complaint", and links them into a graph you can see and search. We explain the mechanics in the reasoning graph post.
Two rules matter more than the rest.
Only you teach your twin. Nothing a visitor says changes how it thinks.
When a shared twin has no recorded view on something, it says so. It can offer an approach, clearly framed as an approach and not your position, and the question lands in your inbox of questions to answer.
What is live, and what is not
Live today: the web app, one shareable public link, the embeddable widget, documents with citations, calibration sessions, the fidelity check and Mirror Score.
Planned, not live: voice, meeting preparation, replies in Slack, Teams, WhatsApp and email, and a marketplace for paid access to twins. We hold no security certifications yet. The comparison page says where other tools are ahead.
The Free tier is permanent: one twin and 50 messages a month, with the first 25 on the Sharp model so you can judge the quality. Details are on the pricing page.