Amaanah: Project Story

About the Project

Amaanah is a voice-first, Shariah-compliant digital finance assistant designed for India's underserved and financially less-literate people. It simplifies KYC using AI voice interaction, explains savings and spending in everyday language, and brings culturally grounded trust to digital banking: one that aligns with both financial dignity and ethical clarity.


What Inspired Us

We were struck by a simple yet powerful insight: “Most fintech is built for those who already trust the system.” But what about those who don’t or can’t?

Millions of Indians, especially from underserved and faith-conscious communities, hesitate to adopt digital finance due to a lack of accessibility, trust, or compliance with their values. Complex UIs, English-only apps, and impersonal onboarding often alienate them before the journey even begins.

Amaanah was born from this gap: to bring empathy, language, and voice into the heart of financial inclusion.

This idea was further established when we went through Antler's NEXT100 list of ideas that will shape the future and found AI automation of financial savings and banking to be a big thing in the future.


What We Learned

  • Simplicity is powerful — voice-first UX reduced cognitive friction dramatically.
  • Users relate better when the assistant speaks to them, not at them.
  • Shariah-compliance isn’t just a feature: it’s a trust-building philosophy. A lot of people of the Islam faith in India and across the globe are unable to open bank accounts as those digital solutions are not shariah compliant. Therefore, it is important to address such a big market and uplift them making them a part of the formal economy. Thereby, providing a big boost to the country too.
  • India's digital users are multilingual by nature : UX must follow suit.

How We Built It

  • Frontend: Built on Bolt for rapid prototyping, mobile-first design, and modular components. We tested our prompts on multiple engines and finally after modifying it using other AI tools and some AI expert's advice online, came up with the single-seamless solution oriented prompt.
  • AI Voice Assistant: Powered by ElevenLabs Voice Agent, embedded directly via custom widget.
  • Multilingual UX: Agent supports voice interaction in Hindi, Urdu, and Hinglish, with future plans for regional extensions.
  • KYC Flow: Simulated PAN, Aadhaar, and selfie verification via conversational AI (scalable to backend APIs).
  • Financial Guidance: Integrated logic for savings tracking, ethical finance nudges, and simplified money insights.

Challenges We Faced

  • Balancing complexity with clarity: Creating a trustworthy KYC process that feels simple without skipping key steps.
  • Voice tuning for local accents: Training agents to handle Hinglish, Tamil and Urdu nuances effectively.
  • Limited screen real estate: Designing a mobile interface that feels rich but faithful. It shouldn't seem lavish, or too "cool" as it might cause hindrance in usage. Therefore, it needed to be simple, multilingual app with normal colors.
  • Privacy: Architecting for data collection that remains transparent, secure, and culturally appropriate.

What’s Next

  • Live integrations with UPI and account aggregators
  • Voice-to-text Aadhaar/PAN entry via Whisper
  • OCR for ID scanning
  • Behavioural nudges for daily savings
  • Offline mode for low-connectivity users
  • Leveraging Federal Bank APIs to build it and make it up and running

Closing Thoughts

Amaanah isn’t just an app. It’s a bridge between trust and technology. For those often left behind, it offers not just access but agency in their own financial lives. We believe this is where inclusion starts, not with features, but with faith.

Built With

  • bolt
  • chatgpt
  • elevenlabs
  • federalbankapis
  • netlify
  • supabase
Share this project:

Updates