I'm Nikhil Jangid, a 2026 B.Tech Computer Science & Engineering graduate from Amity University Rajasthan, based in Jaipur, India.
I build across software engineering, backend systems, full-stack applications, AI products, developer tooling, and C++ problem solving. My strongest interest is the engineering layer where software and AI meet: APIs, data flows, retrieval, model integration, tool use, evaluation, reliability, security, and production-oriented design.
I enjoy going one layer deeper than the framework. When I build a feature, I want to understand the path from request β application logic β database / retrieval β model / service β response, and what happens when something goes wrong.
Understand the system. Build it. Measure it. Improve it.
| π Education | B.Tech CSE Β· 2026 Β· CGPA 8.48 |
| πΌ Recent Experience | Wisflux Tech Labs Β· SDE Intern |
| π§ Problem Solving | 400+ DSA problems in C++ |
| ποΈ Build Focus | Backend Β· AI Systems Β· Full-Stack Β· Developer Tools |
| π Based in | Jaipur, Rajasthan, India |
I prefer projects where the interesting part is not just the UI, but the system behind the product.
Flyeng Career is an AI-powered career guidance and placement-preparation platform I built for students navigating software careers.
The product is designed around a simple problem: students often have plenty of resources but lack a structured way to understand what to learn, which career direction to explore, how to prepare, and what to do next.
I built the platform as a student-facing product with:
- AI-assisted career guidance for more personalized exploration
- Placement-preparation workflows aimed at turning broad preparation into actionable steps
- Conversational and voice-oriented interaction, making the experience more natural than a static resource portal
- A full-stack application connecting the frontend, backend services, database, and AI layer
- A product architecture designed to leave room for richer personalization and AI-driven workflows
Stack: Next.js TypeScript Supabase PostgreSQL Gemini Groq AI APIs
π Live Project
A document question-answering system built around the complete retrieval pipeline:
Documents β ingestion β chunking β embeddings β vector search β retrieved context β LLM generation
The project gave me hands-on experience with the engineering details behind RAG rather than treating retrieval as a black box.
Stack: NestJS PostgreSQL pgvector MiniLM OpenRouter Docker
π Live Project Β· Source Code
A developer tool for AI-assisted code analysis, debugging, complexity reasoning, and algorithm visualization.
It combines an interactive coding environment with AI-assisted analysis to help developers understand code behaviour, identify issues, reason about complexity, and visualize algorithmic concepts.
Stack: Next.js TypeScript Monaco Editor AI APIs
π 1st Place Β· College-level Hackathon Β· January 2026
π Live Project Β· Source Code
A resume engineering application focused on structured editing, AI-assisted improvements, ATS-oriented analysis, and PDF generation.
The goal was to make resume creation a structured engineering workflow rather than simply filling out a template.
Stack: Next.js TypeScript Tailwind CSS Groq Clerk
π Live Project Β· Source Code
Jun 2026 - Aug 2026 Β· Jaipur, India
Worked with NestJS, TypeORM, PostgreSQL, JWT, Docker, background processing, and RAG systems, contributing to backend APIs, transactional workflows, document ingestion, embeddings, and vector retrieval.
The work exposed me to practical concerns around concurrency, data consistency, authentication, background processing, retrieval quality, and backend architecture.
May 2025 - Jul 2025 Β· Remote
Worked on frontend development with modern JavaScript and React-based tooling, gaining practical experience with component-driven interfaces and web application development.
C++ C Java JavaScript TypeScript
React Next.js HTML CSS Tailwind CSS Vite Monaco Editor
Node.js NestJS Express TypeORM PostgreSQL MongoDB Supabase MySQL Firebase OpenAPI Zod
LLM Applications RAG Embeddings Vector Search pgvector Tool Calling MCP Agent Workflows Structured Outputs Context Engineering Model Routing AI APIs
Git GitHub Docker Docker Compose JWT REST APIs Authentication Authorization Background Jobs Testing Observability
I've solved 400+ DSA problems in C++ across LeetCode and GeeksforGeeks.
I don't want problem solving to become a collection of memorized patterns. For each problem, I care about being able to explain:
- Why the approach works
- Why brute force is insufficient
- Time and space complexity
- Edge cases and failure modes
- Why an optimization is valid
- When a familiar pattern does not apply
I also deliberately work on non-standard interview problems, where the harder part is often modelling the problem correctly before choosing an algorithm.
- π₯ 1st Place, College-level Hackathon, January 2026, AI Code Analyzer
- π§ 400+ DSA Problems, C++ Β· LeetCode + GeeksforGeeks
- π» 4,500+ GitHub Contributions
- π GirlScript Summer of Code 2025 Contributor
- π ELUSOC 2026 Participant
- π€ Oracle Agentic AI Certified Foundations Associate, 2026
- π McKinsey Forward Program
LLM applications, RAG, retrieval architecture, tool calling, MCP, structured outputs, context engineering, model routing, and agent workflows.
Evaluation datasets, task-based evaluation, regression testing, failure analysis, tracing, observability, latency, cost, and quality measurement.
Prompt injection, indirect injection, tool permissions, least privilege, sandboxing, secret isolation, untrusted tool output, secure execution, and agent-specific attack surfaces.
Type-safe APIs, NestJS, PostgreSQL, OpenAPI, validation, authentication, authorization, background jobs, concurrency, testing, and API contracts.
MCP servers, coding agents, agent harnesses, AI-native developer tools, SDK generation, model routing, structured outputs, long-running agents, and evaluation infrastructure.
I want to understand what happens underneath the framework: data structures, database queries, network requests, API contracts, model inputs, retrieval steps, and failure paths.
Useful engineering measurements include latency, correctness, retrieval quality, error rates, cost, test coverage, and reproducibility.
Production systems should account for invalid input, unavailable dependencies, inconsistent state, bad model output, unsafe tool requests, timeouts, and partial failures.
Frameworks change quickly. The goal is to understand concepts deeply enough that the knowledge remains useful when the framework, model, database, or tool changes.
I continue strengthening the software engineering fundamentals underneath the AI layer:
C++ Β· Data Structures & Algorithms Β· Databases Β· Computer Networks Β· Operating Systems Β· Backend Architecture Β· Testing Β· System Design
At the same time, I'm interested in where these fundamentals meet newer AI infrastructure:
Evaluation Systems Β· Observability Β· Secure Tool Execution Β· Typed Interfaces Β· Agent Runtimes Β· Retrieval Quality Β· AI-Native Developer Workflows
The long-term goal is to build systems that are not only capable, but understandable, testable, observable, secure, and useful.


