Full Stack Software Engineer focused on building scalable, rule-driven web systems.
- End-to-end web applications using React, Next.js, Node.js, and MongoDB
- Backend systems with authentication, RBAC, schema validation, and structured workflows
- Rule-driven platforms where correctness, edge cases, and fairness matter
- Production-ready APIs with validation, indexing, and error handling
A rule-driven savings platform inspired by real-world informal finance systems.
- Designed contribution windows with late penalty enforcemen
- Modeled MongoDB schemas with compound indexes
- Built REST APIs with authentication (NextAuth) and RBAC
- Implemented idempotent contribution logic
- Designed safe account deletion & governance rules
- Built dashboards for contribution tracking and transparency
- Tech: Next.js (App Router), MongoDB, NextAuth, Tailwind, ShadCN
- 👉 Repo: https://github.com/kaushik0010/KOSH
AI-assisted security monitoring & log analysis system.
- Designed structured LLM-based log triage system
- Implemented Human-in-the-Loop workflows
- Built scalable backend schema for audit tracking
- 👉 Repo: https://github.com/kaushik0010/soc-ai
- REST API Design
- MongoDB Schema Modeling & Indexing
- Authentication & Role-Based Access Control
- Data Structures & Algorithms
- Backend Logic & Edge Case Handling
- AI Integration with Structured Outputs
- Start with unclear problems → write specs → design architecture
- Make trade-offs explicitly
- Ship fast → iterate based on constraints
- Use AI as a multiplier, not a replacement