I released Dossier this week, a Mac app that's like astrology for your texts. You choose a contact, label it personal or work and Dossier creates an insightful and usually hilarious read on your communication style, missed cues, what makes them tick and some icebreakers so you remember what to follow-up on. Dossier also diffs between runs for 'What's New." It's the most fun I've had with texting in a very long time.
I built Dossier to be fun and have a strong point-of-view on the role of AI with personal communication like texts. It redacts PII during analysis. It doesn't store your texts and the server-side analyzer discards all data after generation. It never tells you what to say. it's design to support and deepen, not replace relationships.
A lot of the early users end up sharing the reports with the other person because they're funny, supportive and will make you laugh.
I had been sharing with friends but then OpenAI released their thing. It's worth noting the model provider is Anthropic.
Would especially love HN's feedback on the product and the privacy/security architecture.
I released Dossier this week, a Mac app that's like astrology for your texts. You choose a contact, label it personal or work and Dossier creates an insightful and usually hilarious read on your communication style, missed cues, what makes them tick and some icebreakers so you remember what to follow-up on. Dossier also diffs between runs for 'What's New." It's the most fun I've had with texting in a very long time.
I built Dossier to be fun and have a strong point-of-view on the role of AI with personal communication like texts. It redacts PII during analysis. It doesn't store your texts and the server-side analyzer discards all data after generation. It never tells you what to say. it's design to support and deepen, not replace relationships.
A lot of the early users end up sharing the reports with the other person because they're funny, supportive and will make you laugh.
I had been sharing with friends but then OpenAI released their thing. It's worth noting the model provider is Anthropic.
Would especially love HN's feedback on the product and the privacy/security architecture.