🏆 View Winners 🌟 View Sponsors 👥 All Participants

Mohammad Farhan Alam

Participant

Khalil AI - Accuracy-First Islamic Digital Assistant

Votes0

Score0

🧠Community0×5= 0

🌐Public0×1= 0

Individual

December 2, 2025 at 11:35 PM

Project Demo

Watch Demo

Submission Details

hear-about-us: Linkedin

name: Mohammad Farhan Alam

College / University Name: Other

linkedIn-profile: https://linkedin.com/in/farhanrhine

Project Title: Khalil AI - Accuracy-First Islamic Digital Assistant

Describe what your project does?: Khalil AI is something I built out of my personal issue that I was seeing. mostly today AI assistants often give confident but wrong answers, share fake or broken YouTube links, forget basic details about the user, and respond differently every time. Instead of trying to fix these problems with clever prompts, I decided to fix them properly by redesigning the system behind the assistant. What it does Khalil AI uses LangGraph to control how the assistant thinks and makes decisions. It works with multiple tools, keeps both short-term and long-term memory through SQLite, and uses FAISS-based RAG to make sure answers are grounded in real documents. It also has a strict link-verification layer that checks every YouTube or external link before showing it to the user. On top of that, it comes with 15 useful tools from Islamic knowledge to math, weather, and live web search all wrapped in a clean Streamlit UI running locally through Ollama. Why it matters At the end of the day, people want an AI they can trust. Khalil AI focuses on accuracy, consistency, and verification. Every answer is checked, every link is real, and every response is based on real data instead of guesswork. The whole idea behind Khalil AI is simple: AI shouldn’t just sound smart — it should be correct.

Share a 3–5 minute HD demo video: https://youtu.be/PYmwP4w_6hg?si=utGUWjIt0Zz0JhLO

Hosted Project Link (if any)

GitHub Repository (if any)

artifacts/ supporting material

Is this project your original and proprietary work?: Yes

Does your project include any private, sensitive, or restricted code or content?: No

Do you give us (AI Demos & partners) permission to feature your work on our website and share it with our community (with full credit)?: Yes

Do you consent to us showcasing your name/profile on our social media platforms and YouTube: Yes

Are you open to freelance, internship, or collaborative opportunities based on this project?: Yes, I’m interested

Comments (0)

Explore More

All Participants Hackathon Details

Get In Touch

Connect with Mohammad Farhan Alam to learn more about their project.

LinkedIn Profile

Featured

FutureSmart Agent

A complete AI Agent system for real use cases. RAG, NL2SQL, API actions, multi-agent logic — everything in one place.

Khalil AI is something I built out of my personal issue that I was seeing. mostly today AI assistants often give confident but wrong answers, share fake or broken YouTube links, forget basic details about the user, and respond differently every time. Instead of trying to fix these problems with clever prompts, I decided to fix them properly by redesigning the system behind the assistant. What it does Khalil AI uses LangGraph to control how the assistant thinks and makes decisions. It works with multiple tools, keeps both short-term and long-term memory through SQLite, and uses FAISS-based RAG to make sure answers are grounded in real documents. It also has a strict link-verification layer that checks every YouTube or external link before showing it to the user. On top of that, it comes with 15 useful tools from Islamic knowledge to math, weather, and live web search all wrapped in a clean Streamlit UI running locally through Ollama. Why it matters At the end of the day, people want an AI they can trust. Khalil AI focuses on accuracy, consistency, and verification. Every answer is checked, every link is real, and every response is based on real data instead of guesswork. The whole idea behind Khalil AI is simple: AI shouldn’t just sound smart — it should be correct.