Maheen Ahmed
NUST
· 2026
Email
maheenev22@gmail.com
Phone
923335178330
GitHub
—
Academic
Program
BSCS
CGPA
3.87
Year
2026
Education
SEECS
Address
Islamabad, Pakistan
DOB
—
Career
Current role
—
Target role
—
Skills
Computer Vision, Deep Learning, Object Detection, Segmentation, Synthetic Data Generation, Differentiable Image Modeling, CNNs, Vision Transformers, GANs, Diffusion-inspired pipelines, Backend Development, Full-stack Development, Gaussian Splatting, FastAPI, PostgreSQL, Oracle, Maximo, SAP ERPs
Interests / quote
Google's mission is to organize the world's information. But what if we could orchestrate the world's agentic intelligence?
Verbatim text
The exact text the LLM saw on the page (or the booklet text from the old import).
This is what powers semantic search.
Maheen Ahmed Cell: 923335178330 | Email: maheenev22@gmail.com LinkedIn: https://www.linkedin.com/in/maheenahmed2004/ Address: H.No. 61-C, Block D, North nazimabad , Islamabad , Pakistan PROFESSIONAL PROFILE Computer Science undergraduate with strong research and applied experience in computer vision and deep learning. Has worked in academic lab and international research environments (Summer@EPFL Intern, MITACS Intern @ UAlberta) on object detection, segmentation, synthetic data generation, and differentiable image modeling. Experienced with modern deep learning architectures including CNNs, Vision Transformers, GANs, and diffusion-inspired pipelines, alongside solid backend and full-stack development skills. Comfortable bridging theory and implementation, with hands-on experience in experimentation, model analysis, and building end-to-end AI systems. EDUCATION Computer Science SEECS , Islamabad , 3.87 (2026) INTERNSHIP EXPERIENCE Summer@EPFL 02-Jun-2025 - 01-Aug-2025 -Worked at IVRL headed by Prof. Sabine Susstrunk. -Made the occlusion-based mathematical Dead Leaves model differentiable through gaussian primitives based on 2D/3D Gaussian splatting. -Studied 3D/2D Gaussian Splatting and repurposed gaussians to act as primitives -Studied frequency domain properties of generated images -Experimented with other primitives like triangle splats to better replicate natural image properties. TUKL Deep Learning Lab 03-Jun-2024 - 31-May-2025 - Worked on salient/co-salient object detection pipelines for binary segmentation of rust in wheat crop datasets, under the supervision of Dr.Faisal Shafait. - Explored utilizing salient and co-salient object detection(SOD and COSOD) for wheat rust segmentation on the NWRD/NWRDF datasets -Applied transfer learning to 5 SOTA COSOD models,some transformer-based, and fine-tuned results - Applied architectural changes to best-performing mode ADNOC 01-Jan-2025 - 31-Jan-2026 Received training for using Oracle, Maximo, and SAP ERPs Vision Architech 03-Jun-2024 - 31-Aug-2024 Worked on creating endpoints and setting up the database through FastAPI and PostgreSQL for an AI-powered Analytics web application University of Alberta 01-Jun-2026 - 30-Sep-2026 Will be working on visual motion analysis under the supervision of Dr. Li Cheng. FINAL YEAR PROJECT Decentralized Internet of Agents Google's mission is to organize the world's information. But what if we could orchestrate the world's agentic intelligence? Our project,
AI enrichment
Maheen Ahmed is a Computer Science undergraduate with a 3.87 CGPA, specializing in computer vision and deep learning through internships at EPFL, the University of Alberta, and academic labs. She has hands-on experience with modern architectures like CNNs, Vision Transformers, and Gaussian Splatting, alongside backend development skills using FastAPI and PostgreSQL.
Skills (AI)
["Computer Vision", "Deep Learning", "Object Detection", "Image Segmentation", "Gaussian Splatting", "Vision Transformers", "GANs", "FastAPI", "PostgreSQL", "Python", "Transfer Learning", "Research"]
Status: ai_done