Akash Deep

Loan Approval Rate Dashboard & Document Verification Summary Report

Successfully completed a real-world internship project focused on analyzing loan approval patterns, building meaningful dashboards, and summarizing document verification workflows in a fintech-like use case.

Key Focus Areas

• Analyzed loan approval patterns and translated data into actionable dashboards
• Built structured summaries for document verification workflows
• Strengthened data-driven decision-making and problem-solving approach
• Improved technical communication and reporting clarity

Impact / Outcome

• Bridged engineering fundamentals with applied data analytics
• Reinforced interest in data-driven business decision projects
• Grateful for the guidance and exposure from the AWDH Technology team
• Motivated to deepen skills in data analytics, ML, and scalable systems

Selected for an upcoming summer internship at Accenture.

Designed and developed a full-stack placement management system to automate on-campus recruitment workflows.

Key Highlights

• Implemented server-side eligibility validation for CGPA, active and dead backlogs, academic branch, year gap constraints, and placement tier rules to enforce policy-compliant applications
• Built multi-round hiring workflows with role-based dashboards for Students, Coordinators, and Placement Admins
• Integrated JWT-based authentication, role-based access control, and secure data handling using Prisma ORM
• Added CSV and Google Sheets export for recruiter-ready applicant data and integrated AI-based resume ATS scoring

A single-page attendance manager with subject dashboards, attendance marking, weekly schedule planning, and reporting tools for quick tracking.

Key Features

• Global summary metrics for subjects, units, and overall attendance
• Daily attendance marking with extra class support
• Weekly schedule editor and subject management
• CSV export reports with date-range filtering

A fully offline Android application designed to calculate risk-free betting opportunities using arbitrage strategies. Users can enter back odds, lay odds, and one stake, then instantly compute the corresponding stake and guaranteed profit or loss across all outcomes. Built for speed, simplicity, and real usability, it works reliably even in low-connectivity situations.

Key Features

• Fully offline operation for fast, uninterrupted usage without internet access
• Arbitrage calculation engine that balances bets to minimize or eliminate risk
• Flexible inputs with support for either back stake or lay stake entry
• Real-time profit and loss calculation across possible outcomes
• Floating window support for use on top of other apps during live scenarios
• Clean, no-ads experience with error handling for invalid or incomplete inputs

Tech & Design Focus

• Lightweight and optimized for performance
• Minimal, intuitive UI for quick input during live betting workflows
• Practical validation to keep calculations reliable under pressure

Use Case

Designed for users who actively engage in sports betting and want to exploit arbitrage opportunities efficiently without relying on external tools or websites.

FairVote

Built FairVote, a private election platform that uses Google Sign‑In and Firebase to create secure, eligibility‑based elections with anonymous results. Electors can define candidates, voting limits, and eligible email lists, while voters cast 1..N selections without duplicates. The system preserves privacy by allowing electors to see who voted—but never how they voted—and publishes only aggregated counts on a public results page.

Key Features

• Google authentication for voters and electors
• Eligibility gating via email allowlist
• Configurable candidates and max selections
• Anonymous, aggregated results with a public results view
• Firestore data model for elections, votes, and results