Alora
Your user-controlled memory that syncs across AI tools
38 followers
Your user-controlled memory that syncs across AI tools
38 followers
Alora gives you user-controlled memory across AI tools. Create topics that learn and auto-update as you chat, then inject the right context into any conversation. No more re-explaining - just faster, more accurate answers every time.
👋 Hey Product Hunt, Arjun here.
We built Alora to fix a simple pain: you keep re-explaining yourself to AI, and context gets lost when you switch tools.
What it is
User-controlled memory that syncs across AI tools. Works with ChatGPT and Perplexity today.
How it works
• Create a topic, chat as usual.
• Alora learns and auto-updates in the background.
• With one click, inject the right context into any conversation.
What makes it different
• You can see and edit what AI remembers.
• Per-topic control, not a black box.
• Cross-tool sync, quick setup.
Privacy
No data selling. Full visibility and editability. Data is stored locally in your browser
Would love your thoughts. Which tool should we support next, and what should Alora remember by default? If any part of onboarding feels rough, please tell us so we can fix it fast.
@arjun_dev_singh1 Are you planning an API so other devs can plug Alora’s memory into their own AI tools or custom agents?
@arjun_dev_singh1 @seraermolenko Not at the moment, but it is something we are considering.
such a time saver! Was solving the "over-generalizing" issue difficult? Usually when I try to summarize a large amount of text, the model drops some topics completely
@ashkan_ziaee We're giving users the ability to define how granular they want the topic to summarize to be. We're also allowing users to edit the summary manually so that users can add missing information or fix incorrect memories. Still a lot of iteration needs to happen to solve this issue in a better way
What To Build
Love this user‑controlled memory that travels across AI tools is super useful. Curious how you handle privacy, merging conflicts, and portability between models.
We’re launching today as well if you’re interested, you can take a quick look
@niladri_hazra Thanks! We're storing all data locally in the user's browser. We have a privacy policy on our website. As of now, users are able to see how their topic evolves through version history. Users can manually review what has changed and control exactly what we remember about the topic. We also separate memory away from models into 2 separate layers.
GPT-4o
This is truely awesome! The per-topic control is such a smart idea — I'm *so* tired of AI forgetting everything when I switch tools. Being able to see and edit what it remembers is a game changer, ngl. Kinda genius imo. How's the performance with really long conversations?
Agnes AI
Cool this looks nice! Congrats team! Just curious - how do you balance and manage between long/short -term memory across diff tools?
@cruise_chen Currently, we have a strong emphasis on high-level, long-term memory. Short-term memory is implicit in the current chat window. Also, our tool separates memory away from LLM tools so that they are independent of each other. No difference regardless of which tool you use.
Sellinger AI
Hey,
Great work, just backed you, and we’d love a quick visit to ours.