BlockGuard SLM

Inspiration The increasing number of scams such as phishing, rug pulls and fake transactions in the blockchain world prompted us to take action. The lack of fraud prevention tools motivated us to develop BlockGuard SLM, which aims to protect users and restore trust in the decentralized ecosystem

What we have learned We explored blockchain’s complexities, smart contract vulnerabilities, and real-time fraud detection challenges. This project strengthened our expertise in AI, anomaly detection and decentralized data integration.

How we built it BlockGuard SLM was developed using an AI model trained on blockchain data, smart contract metadata, and historical fraud cases. By integrating decentralized sources such as blockchain nodes, IPFS, GitHub APIs, and crypto forums, we have built a system that can detect fraud and alert in real time

Challenges Faced Key challenges include managing noisy blockchain data, ensuring the accuracy of real-time analytics, and integrating disparate data sources. Overcoming these obstacles helped us develop scalable and effective solutions.

This project demonstrates our commitment to enhance blockchain security and enhance trust through AI-driven innovation.

Built With

• blockchain-explorers-(e.g.
• etherscan-api-for-transaction-details)
• flask-(for-api-backend).-platforms:-ethereum-(for-blockchain-data)
• for
• google-cloud-ai-(for-model-deployment).-databases:-mongodb-(for-transaction-and-metadata-storage)
• ipfs-(for-decentralized-data-storage).-cloud-services:-aws-(for-hosting-and-processing)
• javascript-(for-frontend-and-integrations).-frameworks:-tensorflow
• kubernetes-(for-orchestration)
• languages:-python-(for-ai-and-data-processing)
• metamask
• postgresql-(for-user-data).-apis:-github-api-(to-fetch-smart-contract-metadata)
• pytorch-(for-ai-model-training)
• testing
• transaction
• twitter-api-(for-sentiment-analysis).-other-tools:-docker-(for-containerization)