Deep Bartaria.
Software & AI Engineer
Experience
Artificial Intelligence Intern
- Contributed to an AI-powered semantic search engine for videos. Achieved Precision of 82–88% & Recall of 75–80%.
- Extracted and processed video metadata and transcripts achieving an F1-score of 0.80–0.83.
- Architected multiple SentenceTransformer models for embeddings, improving relevance by 25–35%.
- Scaled the system to handle 10K+ video embeddings with negligible performance degradation for top-5 results.
Software Development Intern
- Developed an end-to-end IPO Web Application using React.js and PostgreSQL, integrating RESTful APIs.
- Achieved average API response latency of ~120–180 ms under normal load.
- Ensured application reliability and scalability via QA testing, improving query execution time by ~30%.
- Prepared technical documentation aligning with fintech compliance, reducing bugs by ~15% during testing.
Selected Works
AI Preparation Platform
LexiPrep
A full-stack platform with real-time video sessions and AI-assisted interview workflows. Deployed 10+ relational models reducing manual data handling by 70%.
Conversational AI & RAG
NoobifyAI
An end-to-end conversational AI assistant solving hallucinations with context-aware memory. Features a 3-zone Streamlit interface with real-time token streaming.
Custom DBMS
Nimbus
A custom DBMS supporting core SQL operations handling up to 1 GB of CSV data. Optimized using Parallel Programming and Threading for low execution time.
Technical Arsenal
Languages
Frameworks & Tools
Databases
Cloud Platforms
System Admin & DevOps
QA & Libraries
Education
Indian Institute of Information Technology Vadodara
Bachelor of Technology in Computer Science & Engineering
Aug 2023 – Present
DBMS, OS, Computer Networks, Software Engineering, Machine Learning, Artificial Intelligence, AI in 5G.
Achievements
INSPIRE Award
Ministry of Science and Technology, Govt. of India
Selected at the state level for innovative scientific ideas.
CSIR Jigyasa Awardee
Council of Scientific and Industrial Research, Govt. of India
Chosen among the top 15 candidates across India.
Certifications
Machine Learning Specialization (Stanford), Introduction to Software Engineering (IBM), Fundamentals of Deep Learning (NVIDIA).
Research
Detecting Concealed Crops in Agriculture using Deep Learning
- Implemented RGB-D deep learning pipeline for detecting visually concealed crops in dense environments.
- Achieved F1-score of ~0.81 and Weighted F-measure (Fβ) of ~0.80+ on the ACOD-12K dataset.
- Improved detection performance by ~10–310% over baseline models (SINet, PFNet, HitNet).
- Designed depth-guided feature fusion and iterative refinement modules.
Let's Build
Something Great.
I am currently open for new opportunities. Whether you have a question or just want to say hi, I'll try my best to get back to you!
Say Hello© 2026 Deep Bartaria. All rights reserved.