Sundar Velkur

About Sundar Velkur

I work at the intersection of structural engineering and machine learning, mainly focused on qualification and reliability of high-performance systems.My background is in structural dynamics, reliability, and uncertainty modeling, and I have spent most of my career developing methods that connect analysis, testing, and real-world performance. A lot of my work involves dealing with complex loading conditions and limited or noisy data.Recently, I developed a sample-size-agnostic ML method for part qualification and damage detection in additive manufacturing (AM) components. I also work with vibration-based methods (LARS) for non-destructive evaluation, and I am correlating these signals with fatigue life and CT-derived porosity.Previously, I have worked on structural reliability, surrogate modeling, and high-dimensional optimization across applications in manufacturing, biomedical systems, and engineering design.In general, I am interested in problems related to structural integrity, fatigue, qualification/validation, and scalable non-destructive evaluation, especially where data-driven methods can complement physics-based understanding.

Experience

1. Sr Mechanical Analysis Engineer Ml-driven am Qualification & Nde

   Metrolaser, Inc

   Apr 2025 — Present · Laguna Hills, CA, US

   Built a sample-size-agnostic ML framework for scalable qualification and damage detection in additive manufacturing (AM) components, eliminating traditional data limitations.2. Established LARS as a non-destructive method for detecting internal defects and process-induced variability, reducing reliance on CT inspection.3. Developing correlation between LARS metrics, fatigue life, and CT-derived porosity to enable faster, data-driven part qualification.

Education

• Indian Institute of Science (IISc)

  Doctor of Philosophy (PhD), Civil Engineering - Structural Engineering

  2008 — 2014

• Birla Institute of Technology and Science, Pilani

  BE (Hons), Civil Engineering

  2002 — 2006