About
I'm a final-year Computer Science student at IIITDM Kurnool with a strong foundation in backend engineering, AI systems, and systems programming. I enjoy working at the intersection of performance and intelligence — building things that are both fast and smart.
Currently I'm interning at Appscrip as a Python/AI Developer, building AI-SEO dashboards, Generative AI ranking systems, and multi-tenant analytics infrastructure. Outside of work I'm developing VBank — a voice-activated banking platform with biometric authentication — as part of a Government of India initiative.
When I'm not coding, I'm competing on LeetCode (top 5% nationally), reading about distributed systems, or building small C++ game engines for fun.
Languages
- Python
- C/C++
- JavaScript
- Shell
Backend
- FastAPI
- Flask
- RESTful APIs
- SQLAlchemy
- Redis
- Celery
AI / ML
- LlamaIndex
- Agentic RAG
- NLP
- Deep Learning
- SpeechBrain
- Google ADK
Frontend
- TailwindCSS
- HTML/CSS
- Alpine.js
Tools
- Docker
- Git
- CI/CD
- Linux
- Postman
Databases
- PostgreSQL
- ChromaDB
- ClickHouse
- SQLite
Experience
-
Jan 2026 — Present Python / AI Developer Intern · Appscrip
Bangalore, India
- ▹ Developed a modular Marketing Analytics & AI-SEO Dashboard using FastAPI, integrating ClickHouse (OLAP) and Supabase (PostgreSQL) to visualize high-volume traffic data.
- ▹ Engineered a Generative AI Ranking System that evaluates website performance across LLMs and chatbots, providing location-specific analytics and actionable optimization suggestions.
- ▹ Implemented Multi-tenancy and RBAC to securely manage distinct organizational environments and isolate client data.
- ▹ Created a competitor benchmarking tool to analyze keyword gaps using Redis for background tasks and daily Cron jobs for automated data fetching.
- FastAPI
- ClickHouse
- PostgreSQL
- Redis
- Generative AI
- RBAC
-
Apr 2024 — Jun 2024 ML Participant · Samsung Innovation Campus
Remote
- ▹ Built a Naïve Bayes classifier achieving 80% accuracy in auto-sorting 10,000+ citizen complaints.
- ▹ Applied NLP pipelines (tokenization, TF-IDF, CountVectorizer), boosting text classification accuracy by 15%.
- ▹ Created data visualizations (confusion matrix, frequency plots) with Matplotlib/Seaborn for datasets exceeding 1 million records.
- ▹ Engineered an optimized A* algorithm, reducing search time by 35% and increasing result relevance by 40%.
- Python
- NLP
- Scikit-learn
- Matplotlib
- A* Algorithm
Achievements
- ★ LeetCode Contest 469 — Rank 1298 / 30,390 2025
- ★ NPTEL Topper — Rank 3 / 13,763 2025
- ★ Student Excellence Award — Rank 1 / 320 · IIITDM 2023
- ★ JNV Selection — Top 80 / 4,000 2013
Projects
-
VBank — Voice-Based Banking Platform
A secure, voice-activated banking system with real-time multimodal biometric authentication enabling hands-free transactions for accessibility.
- FastAPI
- SpeechBrain
- InsightFace
- JWT
- PostgreSQL
- CI/CD
- Python
-
Scheme-Saathi — AI Public Service Platform
A multilingual AI-powered platform helping 1,000+ citizens discover and access government schemes through natural language search and personalized recommendations.
- Flask
- SQLAlchemy
- Redis
- Celery
- ChromaDB
- TailwindCSS
- Alpine.js
- OAuth2
-
RAG-Based Agentic Assistant
A production-grade retrieval-augmented generation system supporting 1,000+ heterogeneous documents with agentic and non-agentic pipelines, built on LlamaIndex and multiple LLMs.
- LlamaIndex
- FastAPI
- PostgreSQL
- ChromaDB
- Gemini 2.5
- LLaMA 3.2
- DeepSeek
- Python
-
TiniMilitia-Shooter — 2D Game Engine
A high-performance 2D shooter engine in modern C++17 using OpenGL and GLFW, sustaining 60 FPS with 100 active entities via RAII resource management and deterministic game loops.
- C++17
- OpenGL
- GLFW
- CMake
- RAII
- OOP
Writing
-
1 April 2025
What I Learned Building a Production RAG System
Lessons from building a RAG pipeline that handles 10,000+ retrieval requests — covering chunking strategies, reranking, agentic pipelines, and the LLM benchmarks that actually matter.
- RAG
- LlamaIndex
- Python
- LLMs
- FastAPI