AI Engineer
AI Engineer with hands-on experience designing and deploying LLM-powered applications, Retrieval-Augmented Generation systems, and machine learning solutions using LangChain, LLaMA 3, PyTorch, FastAPI, and AWS.
NV-ITECH Security Systems and Equipment Trading LLC
Nov 2025 – Apr 2026 · Dubai
Assisted in designing and developing AI/ML algorithms and models; built data preprocessing and feature engineering pipelines; developed, optimized, and evaluated machine learning and deep learning models; supported production-ready AI solution implementation and testing in collaboration with cross-functional teams and stakeholders.
NV-ITech Security Systems and Equipment Trading L.L.C.
6 mos · Dubai, United Arab Emirates
Worked on internal data and machine learning projects, managing the full pipeline from raw data to model evaluation. Performed exploratory data analysis on internal business datasets to surface trends, anomalies, and data quality issues that informed modelling and feature selection decisions. Built preprocessing and feature engineering pipelines that converted raw business data into clean, ML-ready inputs. Developed and evaluated machine learning models across multiple business use cases, contributing to model selection, hyperparameter tuning, and performance benchmarking to identify the best-fit solutions.
M.Tech, Artificial Intelligence
2023 – 2025
B.Tech, Computer Science Engineering
2019 – 2023
Master of Technology - MTech, Artificial Intelligence
2023 - 2025
Built a conversational HR assistant using LangChain, FAISS, and LLaMA 3.1 (8B) for context-aware question answering across 1,470 HR records. Designed retrieval workflows and prompt-engineering strategies, and developed an evaluation framework across 20 benchmark queries, achieving 100% retrieval success.
Built a machine learning-based phishing detection system using Python and Scikit-learn to classify malicious URLs and messages. Developed and containerized a FastAPI inference service with Docker, and implemented experiment tracking, model versioning, and performance monitoring using MLflow and DagsHub.
Fine-tuned a Pegasus transformer model using PyTorch and Hugging Face Transformers on GPU-accelerated hardware for abstractive dialogue summarization. Built an end-to-end NLP pipeline with ROUGE evaluation and deployed a FastAPI inference service.
This project focuses on leveraging advanced Natural Language Processing and deep learning techniques to identify signs of depression from social media text data. It combines BERT-based language representations with machine learning and graph-based models, such as Graph Attention Networks (GAT), to capture both semantic and contextual patterns in user-generated content. The goal is to support early mental health assessment through scalable and intelligent text analytics.
IEEE · Apr 23, 2025
IEEE · Apr 23, 2025
Springer · Jul 4, 2024