I'm a Computer Engineering student at VJTI, building toward a career in backend engineering, distributed systems, and applied machine learning. I like pairing algorithmic rigor — 650+ problems solved across LeetCode, Codeforces, and CSES — with production-grade engineering: normalized relational schemas, low-latency APIs, and agentic RAG pipelines.
- 🔭 Building full-stack systems with React, Next.js, FastAPI & PostgreSQL
- 🧠 Exploring RAG pipelines, vector search, and LLM orchestration with LangChain & ChromaDB
- 🌱 Sharpening distributed systems & concurrent programming fundamentals through coursework and projects
- 🤝 Active CP Club Member, Community of Coders (CoC), VJTI
- 💼 Open to: Software Engineering Internships — Full-Stack / Backend / AI-ML
- 📫 Reach me at tahavaliji52@gmail.com
Languages
Also comfortable with SQL for relational schema design.
Frontend
Backend & Databases
Cloud, DevOps & Tooling
TPQA Management System — React, Next.js, PostgreSQL, FastAPI
A full-stack management platform focused on query performance and data integrity at scale.
| Stack | React, Next.js, FastAPI, PostgreSQL |
| Scale | Dual-layer server + client-side pagination across all relational entities |
| Performance | 50%+ reduction in API response times under concurrent load |
| Security | 3NF/BCNF-normalized schema — zero referential integrity violations under load testing |
| Impact | Eliminated data redundancy across the entire schema via composite SQL indexing |
| Repository | Once-1296/dbms_minmax_proj_1 |
Engineered a dual-layer REST API pagination architecture (server-side + client-side) using FastAPI and Next.js, then backed it with a strict 3NF/BCNF-normalized PostgreSQL schema and composite indexing to keep queries fast as data grows.
A* Pathfinding Visualizer — Python, Pygame, Data Structures & Algorithms
An interactive visualizer demonstrating provably optimal pathfinding in real time.
| Stack | Python, Pygame |
| Scale | Tested across 100+ user-defined grid configurations |
| Performance | Optimal O(b^d) shortest-path computation via A* with a Min-Heap and O(1) hash set |
| Security | N/A — local visualization tool |
| Impact | Consistent optimal-path convergence, verified through worst-case node-expansion analysis |
| Repository | Link pending — send it over and I'll wire it in |
Implemented A* search with an admissible Manhattan Distance heuristic, a Min-Heap for frontier management, and an O(1) hash set for visited-node lookups, wrapped in a dynamic Pygame grid for real-time visualization.
AI-Powered Device Health Monitor — PyQt5, Streamlit, Supabase, ChromaDB, RAG
A distributed monitoring platform with autonomous, RAG-driven anomaly diagnosis.
| Stack | PyQt5, Streamlit, Supabase, ChromaDB, LangChain |
| Scale | Distributed multi-agent platform with concurrent multi-user access |
| Performance | Sub-second dashboard refresh latency |
| Security | Google OAuth session enforcement for concurrent multi-user access |
| Impact | Autonomous anomaly diagnosis via an agentic RAG pipeline correlating live telemetry |
| Repository | MohdHedayati/System-Health-Metric-Analyser |
Built an agentic RAG pipeline over ChromaDB that correlates live telemetry to autonomously flag anomalies, backed by a Supabase backend with edge-based ML inference and Google OAuth for secure, concurrent multi-user access.
Mentee — AI/ML Systems Monitoring · Inheritance, Community of Coders (CoC), VJTI
Oct 2025 – Present
Contributed to Inheritance 2025 as a mentee, completing an AI/ML system monitoring project, while staying active as a CP Club member supporting the club's problem-solving culture.
- Completed an AI/ML system monitoring project under the Inheritance mentorship track
- Active CP Club Member, contributing to community problem-solving sessions
"Turning algorithms into architecture — one commit at a time."
