Muhammad Tayyab Iftikhar

PROFESSIONAL PROFILE Computer Science undergraduate at NUST with a focus on Large Language Models and generative AI. Experienced in fine-tuning LLMs, building RAG pipelines, and developing scalable AI applications. Skilled in machine learning, deep learning, and full-stack application design, translating technical requirements into production-ready solutions. Seeking an AI/ML role to create LLM-driven and generative AI systems with real-world impact. EDUCATION BSCS SEECS , Islamabad , 3.29 (4) INTERNSHIP EXPERIENCE ESCASA 09-Jun-2026 - 15-Aug-2026 Trained deep learning models for text-to-speech synthesis and image caption generation. Developed and evaluated model pipelines for accurate and robust outputs. HPC NUST 04-Aug-2025 - 25-Jan-2026 Developed RAG pipelines and fine-tuned large language models, exploring Ollama and Hugging Face ecosystems, and currently working on the NUST AI help bot. Built an interactive system allowing students to query and explore models in real time. Machvis NUST 04-Mar-2025 - 25-Jan-2026 Conducted exploratory data analysis on Lahore population and air pollution datasets, generating insights and visualizations. Later, worked on UAV imagery and machine learning to analyze wheat genotypes, preprocessing data and extracting structural, spectral, and texture features for phenological analysis. FINAL YEAR PROJECT Analyzing UAV based multispectral data for genotypic aware crop performance Assessment Modern agriculture increasingly uses UAV-based multispectral imagery to monitor crop performance. However, current methods often treat entire fields as homogeneous crop populations, overlooking the underlying genotypic or varietal diversity present in experimental or breeding fields. Each plot in such fields typically hosts a specific genotype or variety, often replicated across spatial and temporal dimensions. Without accounting for these varietal differences, analysis of phenological stages, biomass estimation, fractional cover, and other morphological / physiological features lack granularity and cannot inform genotypic performance comparison or variety selection. This project aims to bridge the gap identified in the above problem statement by performing a detailed, genotypic-aware analysis of UAV-collected multispectral crop data across multiple seasons and crops (Sunflower and Wheat) to identify and quantify performance variability, cluster high-performing varieties, and provide trends & insights for plant breeders and agronomists through a well-designed dashboard / interface. TECHNICAL EXPERTISE Data Engineering & Feature Engineering Processed, cleaned, and transformed datasets for machine learning and deep learning models. Worked with structured,

Computer Science undergraduate at NUST with a focus on Large Language Models and generative AI. Experienced in fine-tuning LLMs, building RAG pipelines, and developing scalable AI applications. Skilled in machine learning, deep learning, and full-stack application design, translating technical requirements into production-ready solutions. Seeking an AI/ML role to create LLM-driven and generative AI systems with real-world impact.