| title | BioScholar AI Unified |
|---|---|
| emoji | 🔬 |
| colorFrom | blue |
| colorTo | green |
| sdk | docker |
| app_port | 7860 |
| Platform | URL | Status |
|---|---|---|
| 📊 Interactive Deck (Live) | https://biomed-scholar.web.app/presentation/index.html | ✅ Live |
| 🌐 Web App (Firebase) | https://biomed-scholar.web.app | ✅ Live |
| ⚡ Backend API (HF Space) | https://sasidhara123-biomed-scholar-api.hf.space | ✅ running |
| 🤖 Maverick Telegram Bot | https://t.me/Meverick_AI_bot | ✅ Active |
| 💠 Bot Landing Page (HF) | https://sasidhara123-maverick-ai-bot.hf.space | ✅ Live |
| 🐙 GitHub Repository | https://github.com/sasi123-stack/BioScholar-AI | ✅ Public |
| 🔥 Firebase Console | https://console.firebase.google.com/project/biomed-scholar | 🔒 Private |
A dedicated analytical hub for real-time biomedical discovery tracking:
- Fast-Emerging Topics: Pulse-tracking of high-growth research areas (e.g., GLP-1, mRNA).
- Sentiment Pulse: AI-driven community sentiment analysis on research safety and efficacy.
- Global Field Breakdown: Visualizing the distribution of publications across Oncology, Immunology, etc.
- Live Discovery Feed: Real-time ticker of major breakthrough papers within the last 24 hours.
Expanded search universe beyond peer-reviewed literature:
- Serper API Integration: One-click toggle to fetch real-world news, clinical trial updates, and public health reports from the World Wide Web.
- Unified Results: Web results are seamlessly blended with PubMed/ClinicalTrials data with functional source linking.
Implemented a high-availability search architecture to ensure zero downtime:
- Primary Engine: High-speed semantic search via Bonsai Elasticsearch.
- Failover Engine: Real-time integration with NCBI Entrez API.
- Auto-Switching: The system automatically detects upstream provider outages and reroutes queries to ensure consistent delivery.
The centerpiece of the Research Desk, providing real-time evidence synthesis:
- Llama 3.3 Powered: High-speed, high-accuracy RAG synthesis via Groq.
- Maverick Insight Box: Automatically generates a structured "Research Briefing" above search results.
- Context-Aware Linkage: Seamless navigation between Research Desk, Maverick Bot, and Research Trends.
- Dynamic Source Mapping: Identifies and links citations across all indexed sources.
A professional, cinematic web-based slide deck for the April 10th showcase:
- Premium Design: Space Grotesk typography & Glassmorphism UI.
- Presenter Notes: Fast-access one-liners for a polished 10-minute delivery.
- 3D Transitions: Seamless navigation between architectural and feature pillars.
| Command | Description |
|---|---|
/start |
Welcome message & quick start |
/search <topic> |
AI biomedical literature search |
/history |
View recent conversation |
[PDF/Image] |
Direct synthesis of uploaded documents |
┌─────────────────────────────────────────┐
│ BioMedScholar AI Stack │
├─────────────────┬───────────────────────┤
│ Frontend │ Firebase Hosting │
│ (HTML/CSS/JS) │ biomed-scholar. │
│ │ web.app │
├─────────────────┼───────────────────────┤
│ Backend API │ HF Space (Docker) │
│ (FastAPI + │ biomed-scholar-api │
│ Entrez/BS) │ │
├─────────────────┼───────────────────────┤
│ Telegram Bot │ HF Space (Docker) │
│ (python- │ maverick-ai-bot │
│ telegram) │ │
└─────────────────┴───────────────────────┘
| Layer | Technology |
|---|---|
| Frontend | HTML5, CSS3, Vanilla JS |
| Backend | FastAPI (Python 3.10) |
| Search | Bonsai Elasticsearch, NCBI Entrez, Serper API |
| AI Inference | Groq & Chrome Gemini Nano (On-Device) |
| Hosting | Firebase & Hugging Face Spaces |
| Visualization | D3.js v7 & CSS Custom Properties |
- Feature: Standalone interactive presentation app in
/presentation. - Outcome: Optimized for the April 10th 10-minute presentation at 3:20 PM IST.
- Feature: On-device "Local AI Summary" container in article details.
- Benefit: Zero-latency research synthesis for Chrome users.
- Status: Removed all traces of "BETA" labels across UI/UX.
- Logic: Consolidated Trends CSS (24px radius, 28px padding) for a premium production feel.
We have implemented a high-fidelity Selenium-based Showcase Script (selenium_test.py) for automated product demos and QA verification.
- Prepare Environment:
pip install selenium webdriver-manager
- Execute Script:
python selenium_test.py
- Wait for Playback: The script will automatically navigate the v1.6.5 platform, trigger live analysis, execute searches, and highlight the Local AI summaries. Perfect for creating promotional videos with voiceovers.
- 🎥 Official Product Showcase: April 9th, 2026 | 3:40 PM IST
- 📊 Interactive Deck (Live): https://biomed-scholar.web.app/presentation/index.html
- 📊 Interactive Deck (Local):
/presentation/index.html