The Milestone: 1st Position in the Internal Screening Round
Participating in the Smart India Hackathon (SIH) is one of the most exciting and demanding experiences for any engineering student in India. Before getting the chance to represent our institution at the national level, we had to clear the first big hurdle: the College Internal Hackathon Screening Round.
Our team spent days refining our presentation, understanding the core requirements, and preparing a rock-solid technical blueprint. When the results were officially announced, we were thrilled to have secured the 1st position among all the competing teams in our college!
Winning the internal round validated our preparation and gave our team immense confidence, but we knew that the real challenge had only just begun.
The Biggest Hurdle: Selecting the Perfect Problem Statement
Once the internal round was cleared, our biggest struggle began: choosing the right Problem Statements (PS) from hundreds of problem statements released by various government ministries and departments.
Selecting a problem statement isn't just about picking what looks interesting—it requires balancing several critical factors:
- Technical Feasibility & Scope: Ensuring the problem is challenging enough to stand out while being achievable within hackathon deadlines.
- Societal & Industry Impact: Addressing a real-world bottleneck where technology creates a measurable, positive impact.
- Team Strengths Alignment: Leveraging our combined expertise in full-stack web engineering, scalable databases, APIs, and modern system architectures.
- Clear Evaluation Criteria: Identifying problems with well-defined deliverables that jury panels can rigorously evaluate.
The Final Decisions: SIH26102 and SIH26097
After multiple rounds of team discussions, debate sessions, and architectural feasibility checks, we narrowed down our focus and finalized two standout problem statements that tackle critical national challenges:
1. SIH26102: AI-Powered Anomaly & Fraud Detection for MPLAD Scheme
- Official Statement: Development of an AI-powered system to detect anomalies, fraud, and inefficiencies in MPLAD Scheme implementation regd.
- The Core Problem: The Members of Parliament Local Area Development Scheme (MPLADS) handles substantial public funds distributed across thousands of local infrastructure projects. Tracking fund utilization, identifying duplicate projects, catching fraudulent contractors, and eliminating bureaucratic delays across districts is historically difficult with manual oversight.
- Our Solution & Technical Approach:
- Machine Learning Anomaly Detection: Implementing unsupervised clustering and anomaly detection models to identify abnormal fund expenditures, project completion delays, and inflated cost estimates.
- Cross-District Duplicate Detection: Natural Language Processing (NLP) to detect overlapping project proposals submitted across adjacent jurisdictions.
- Transparent Analytical Dashboard: Building a clean, real-time analytics portal with geospatial project mapping and automated risk scores for auditing authorities.
2. SIH26097: AI-Driven Multilingual Voice Assistant for PM-AJAY
- Official Statement: AI-Driven voice Assistant for livelihood Mapping and NSQF-Aligned Skilling Recommendations for SC Communities under GIA component of PM-AJAY.
- The Core Problem: Under the Grants-in-Aid (GIA) component of the Pradhan Mantri Anusuchit Jaati Abhyuday Yojana (PM-AJAY), connecting beneficiaries from Scheduled Caste (SC) communities with skill development programs is hindered by digital illiteracy, language barriers, and complex application portals.
- Our Solution & Technical Approach:
- Multilingual Conversational AI: Integrating high-accuracy Automatic Speech Recognition (ASR) and Text-to-Speech (TTS) models supporting regional Indian languages and local dialects.
- Livelihood & Skills Mapping: An intelligent recommendation engine that analyzes existing skills, educational backgrounds, and local market demands to recommend NSQF-aligned (National Skills Qualifications Framework) training programs.
- Accessible & Low-Bandwidth Interface: Creating an intuitive voice-first interface that functions reliably on basic smartphones with voice prompts and automated enrollment assistance.
Team Brainstorming & Architecture Blueprint
With both problem statements selected, we immediately jumped into planning and architecting the solutions:
- System Architecture: Mapping out data models, database schemas, and microservice/monolithic service boundaries.
- UI/UX Wireframing: Creating clean, distraction-free mockups to guarantee exceptional usability during evaluation.
- Milestone Tracking: Dividing responsibilities across frontend, backend, API integrations, and documentation.
Next Steps on the Road to the Grand Finale
With our problem statements locked in and foundational architecture designed, our roadmap for the upcoming weeks includes:
- Developing the Core MVP: Building out the functional prototypes for both selected problem statements.
- Conducting Rigorous Testing: Stress-testing edge cases, data validation, and security layers.
- Iterating with Mentor Feedback: Gathering continuous feedback from faculty guides and industry mentors to polish our submissions.
Stay tuned for more updates as we continue building and progressing through our Smart India Hackathon journey!
