Help & Supportcareers@gravityer.com
ML Engineer with 2+ years in AI systems, data pipelines, and backend infrastructure.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
ML Engineer with experience across Al systems, large-scale data pipelines, and backend infrastructure. I hold an MScAC from the University of Toronto and work at J-Squared Technologies, building real-time multimodal systems that run reliably under hardware, latency, and privacy constraints. My focus spans Vision-Language Models, self-supervised learning, and dynamic model adaptation.
University of Toronto
MSc: Applied Computing
September 1, 2024 – December 1, 2025
King Fahd University of Pet. & Minerals (KFUPM)
BSc: Computer Science
August 1, 2018 – June 1, 2023
J-Squared Technologies
Machine Learning Engineer
January 1, 2026 – Present
Toronto, Ontario, Canada
J-Squared Technologies
Machine Learning Engineer, MScAC Industry Cooperation
May 1, 2025 – December 1, 2025
Toronto, Ontario, Canada
University of Toronto
Teaching Assistant — CSC207 Software Design
September 1, 2024 – May 1, 2025
Toronto, Ontario, Canada
stc
Data Solutions Developer
September 1, 2023 – September 1, 2024
Riyadh, Riyadh Region, Saudi Arabia
Solutions by stc
Backend Java Developer (Internship)
June 1, 2022 – January 1, 2023
Riyadh, Riyadh Region, Saudi Arabia
4D Gaussian Splatting | Learned Pruning for Dynamic View Synthesis
June 17, 2026 – Present
Introduced a data-driven pruning framework for 4DGS combining importance scoring and temporal gating. Eliminated up to 77% of Gaussians across five benchmarks, reducing memory/compute requirements with no impact on quality.
LLM Unlearning | Gradient-Based Methods on LLaMA 3.1
June 17, 2026 – Present
Evaluated gradient-based unlearning on LLaMA 3.1 8B using LoRA; identified critical flaws in the TOFU benchmark's R_truth metric and improved gradient orthogonal projection by anchoring retain gradients to the pre-forgetting model state.
Multimodal Fusion | Benchmarking VLM Architectures
June 17, 2026 – Present
Tested ensemble/fusion strategies (BERT + ViT) on CUB-200-2011; middle fusion matched ensemble performance at half the cost.
Future Geniuses Scholarship
Dallah Albaraka
June 1, 2026 – Present
Cultural Fit Analysis
The candidate's diverse project portfolio, ranging from academic research in advanced ML topics (4DGS, LLM Unlearning, Multimodal Fusion) to industry experience in real-time edge AI and large-scale data solutions, indicates a broad intellectual curiosity and adaptability. Their academic achievements and scholarship suggest a drive for excellence. The experience as a TA and working in Agile teams points to a collaborative spirit. The blend of research-oriented projects and practical industry applications aligns well with a dynamic, innovation-focused culture.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving and research skills through their academic projects, identifying flaws in benchmarks and proposing novel solutions. Their role as a Teaching Assistant indicates good communication and mentorship abilities. Experience in Agile environments and coordinating with various teams suggests an ability to work effectively in structured development settings. The focus on real-time edge AI systems and optimizing for hardware/latency constraints shows a practical, performance-oriented mindset.