Latasha Singh

B.Tech Computer Science graduate · Aspiring AI/ML Engineer

New Delhi, India

@latasha_singh

LinkedIn GitHub

0 followers

About

Computer Science graduate with project and research experience in Machine Learning, Deep Learning, and Generative AI. Built RAG and medical imaging models using Python, PyTorch, TensorFlow, FastAPI, and cloud/data tools, with experience in data analysis and dashboarding.

Experience

AI/ML Intern

National Institute of Technology Delhi (NIT Delhi)

Feb 2026 – Aug 2026

Designed a novel attention-based deep learning architecture for multi-class lumbar spine MRI segmentation; benchmarked against seven state-of-the-art models; validated results through ablation studies and statistical significance testing; resolved reproducibility issues across PyTorch training pipelines.
Data Analyst Trainee

ThinkNEXT Technologies Pvt. Ltd.

Jun 2025 – Aug 2025

Collected, cleaned, and transformed 50,000+ records; designed an interactive Power BI dashboard; generated insights and presented findings to stakeholders.

Education

2026

B.Tech

Atal Bihari Vajpayee Govt. Institute of Engineering and Technology

Awarded by HPTU

Computer Science and Engineering (CSE) · 2022 – 2026

Computer Science and Engineering (CSE) · 2022 – 2026

Skills

<3 months

C SQL CSS HTML Java UNET Keras Faiss Python OpenCV FastAPI PyTorch Resnet-50 TensorFlow Embeddings scikit-learn Vector Search Responsible AI CBAM knowledge Semantic Search Prompt Engineering Large Language Models Convolutional Neural Networks Retrieval Augmentation Generation Generative AI (ChatGPT, Copilot, Gemini)

Tools / apps / platforms

3 to 6 months

Microsoft Excel Microsoft Power BI

<3 months

Git MySQL GitHub Google Cloud Platform

Languages

Fluent

Hindi Speak English Speak

Projects

OTT Platform Analytics Dashboard

Power BI, Python, SQL

Analyzed 50,000+ Netflix/Amazon Prime records and built a 3-page interactive Power BI dashboard on genre trends and platform performance.
Bone Structure Enhancement from Chest X-Rays

U-Net, VGG19, TensorFlow

Designed a deep learning pipeline using U-Net with VGG19-based perceptual loss to enhance bone structure visibility in DRR chest X-ray images and evaluated output quality using SSIM and PSNR.
SpineAI - Lumbar MRI Segmentation

UNet++, EfficientNet B4, PyTorch, Streamlit

Developed a multi-class vertebral structure segmentation model achieving a Dice score of 0.89 and IoU of 0.82 on the SPIDER dataset; deployed a Streamlit inference application with automated clinical measurement extraction and PDF report generation.
DocuChat - RAG-Based Document Q&A System

Python, FastAPI, FAISS, Sentence Transformers, Groq LLM, Streamlit

Built a RAG pipeline for grounded PDF question answering using document parsing, chunking, sentence transformer embeddings, and FAISS vector search; implemented cited-source response generation; deployed as a FastAPI service with a Streamlit demo app.

⚽ Extracurricular activities

Club Head, Cultural Club

Leadership role in a cultural club.
Hackathon 2025 Organizer

Organized a hackathon event.
NSS Volunteer

Supported event planning, logistics, and community service.

Explore the Broxer community

💬 Q&A Ask about roles, salaries and interviews — and help others. 👥 Circles Join discussions on skills, companies and industries. 📰 Feed Follow updates and insights from the community.