[{"content":" # title: \u0026ldquo;Home\u0026rdquo; # ","date":"1 January 2025","externalUrl":null,"permalink":"/","section":"","summary":"","title":"","type":"page"},{"content":" About Me # I\u0026rsquo;m a Python developer and Machine Learning Engineer with over 3 years of experience building algorithms in both research and industry. I have a strong foundation in applied mathematics — probability, statistics, data analysis, and machine learning — and hands-on experience handling large datasets using tools like PostgreSQL and MongoDB.\nI hold an MSc in Mathematics from the Technical University of Munich, where I specialised in Algebra and Geometry, and a BSc in Mathematics from Shiv Nadar University, where I graduated with Distinction.\nI am currently based in Munich and open to new opportunities in ML engineering, data science, and Python development.\nEducation # Technical University of Munich 2021 – 2024 M.Sc. Mathematics Specialisation in Algebra and Geometry. Coursework in Deep Learning, General Relativity, and Optimization. **Thesis:** Typical Entanglement in Fermionic Gaussian States **Supervisor:** Prof. Dr. Robert König Shiv Nadar University 2017 – 2021 B.Sc. Mathematics Graduated with Distinction. **Thesis:** Characteristic Classes of Vector Bundles **Supervisor:** Prof. Dr. Amber Habib Languages # Language Level Tamil Native English Fluent German Fluent Hindi Fluent Interests # Football — competitive level Violin — Grade 5, Trinity College London (with Distinction); Grade 2 Music Theory Fencing — preparing for tournament qualification, Deutscher Fechter-Bund ","date":"1 January 2025","externalUrl":null,"permalink":"/about/","section":"","summary":"","title":"About","type":"page"},{"content":" Get in Touch # I\u0026rsquo;m currently open to new opportunities in ML engineering, Python development, and data science. Feel free to reach out.\nEmail arav.erode@gmail.com LinkedIn aravind-srini GitHub Weierstrash Location Munich, Germany ","date":"1 January 2025","externalUrl":null,"permalink":"/contact/","section":"","summary":"","title":"Contact","type":"page"},{"content":" Work Experience # uNaice GmbH Feb 2026 – Mar 2026 Backend Developer (Fixed-term) Bremen (Remote) - Contributed to Python-based backend services (REST/GraphQL APIs) and machine learning components. - Supported data integration and processing pipelines (JSON, CSV, XML). - Implemented an RSS feed scraper as a new feature for an existing API. - Collaborated with frontend developers to ensure API stability and performance. Kreatory GmbH Sep 2024 – Aug 2025 Machine Learning Engineer Munich - Developed and fine-tuned image classification models for influencer marketing applications. - Implemented ML algorithms for image clustering using PCA and statistical methods. - Extracted and prepared data from MongoDB for analysis pipelines. - Supported deployment of ML models on AWS. Technical University of Munich Jul 2022 – Aug 2024 Research Assistant – Chair of Engineering Geodesy Munich - Developed Python algorithms for optimisation and geometric modelling (solar panels, building layouts). - Built data pipelines using AprilTags for 3D building measurement. - Co-authored multiple peer-reviewed scientific publications. Certifications # University of Illinois Urbana-Champaign Oct 2024 Object-Oriented Data Structures in C\u0026#43;\u0026#43; (via Coursera) - Fundamentals of C++. - In-depth study of pointers and memory management. - Classes and other data structures. ","date":"1 January 2025","externalUrl":null,"permalink":"/experience/","section":"","summary":"","title":"Experience","type":"page"},{"content":" Publications # 2025 # Integrated Workflow for a Semi-Automated and Robotic Facade Renovation K. Iturralde, W. Shen, R. Bazan, O. Amo Grau, …, A. Srinivasaragavan, S. Das, T. Bock, C. Holst Frontiers in Built Environment, vol. 11, 2025. ISSN: 2297-3362. DOI: 10.3389/fbuil.2025.1649278\nA Robust Pipeline for Facade Geometry Estimation Using AprilTags and Digital Imaging A. Srinivasaragavan, D. Liu, K. Iturralde, C. Holst Proceedings of the 42nd International Symposium on Automation and Robotics in Construction (ISARC), Montreal, Canada, Jul. 2025, pp. 1530–1535. ISBN: 978-0-6458322-2-8. DOI: 10.22260/ISARC2025/0199\n2024 # Enhanced Precision in Built Environment Measurement: Integrating AprilTags Detection with Machine Learning S. Tan, A. Srinivasaragavan, K. Iturralde, C. Holst Proceedings of the 41st International Symposium on Automation and Robotics in Construction (ISARC), Lille, France, Jun. 2024, pp. 1295–1298. ISBN: 978-0-6458322-1-1. DOI: 10.22260/ISARC2024/0167\n2023 # An Automated Prefabricated Facade Layout Definition for Residential Building Renovation K. Iturralde, S. Das, A. Srinivasaragavan, T. Bock, C. Holst Buildings, vol. 13, no. 12, 2023. ISSN: 2075-5309. DOI: 10.3390/buildings13122981\n","date":"1 January 2025","externalUrl":null,"permalink":"/publications/","section":"","summary":"","title":"Publications","type":"page"},{"content":" Technical Skills # Languages \u0026amp; Frameworks # Skill Proficiency Python ●●●●● C++ ●●●○○ Matlab ●●○○○ Data \u0026amp; Databases # Skill Proficiency PostgreSQL ●●●●○ MongoDB ●●●○○ NumPy ●●●●○ Machine Learning \u0026amp; AI # Skill Proficiency Scikit-learn ●●●●○ TensorFlow ●●●●○ Computer Vision ●●●●○ MLFlow ●●●○○ Infrastructure \u0026amp; Tools # Skill Proficiency AWS ●●●○○ Docker ●●●○○ Git ●●●●○ MS Office ●●●●○ Areas of Expertise # Computer Vision — image classification, object detection, AprilTag-based pipelines ML Engineering — model development, fine-tuning, deployment on AWS Data Engineering — pipeline design, data extraction and preparation, large dataset handling Applied Mathematics — probability, statistics, linear algebra, optimisation, quantum information Backend Development — REST/GraphQL APIs, data integration pipelines ","date":"1 January 2025","externalUrl":null,"permalink":"/skills/","section":"","summary":"","title":"Skills","type":"page"},{"content":"I am planning to publish posts in the area of Mathematics, Applied Mathematics and Programming.\n","date":"14 August 2023","externalUrl":null,"permalink":"/posts/","section":"","summary":"","title":"","type":"posts"},{"content":"When people think about Machine Learning, they often imagine complex neural networks and massive models. But in practice, some of the most powerful tools come from classical mathematics and statistics. One of them is Principal Component Analysis (PCA).\nI use PCA regularly because it sits perfectly at the intersection of:\nLinear algebra Statistics Practical data science And it solves very real problems.\nWhat problem does PCA solve? # In many ML projects, especially in computer vision and data analysis, we work with:\nHigh-dimensional data Noisy features Redundant information PCA helps by:\nReducing dimensionality Removing correlations between features Making data easier to visualize and process Speeding up downstream models Mathematically, PCA finds new axes (principal components) that maximize variance. Practically, it gives us simpler and cleaner data.\nA simple intuition # Imagine you have 100 features, but most of the information actually lives in just 10 meaningful directions.\nPCA finds those 10 directions.\nInstead of learning on 100 dimensions:\nYou learn on 10 that really matter.\nThis is both elegant and extremely useful.\nHow I used PCA in real projects # In my work on image clustering and classification:\nPCA helped reduce feature size before clustering Improved performance of downstream ML models Made visualization of image embeddings much clearer Reduced noise from irrelevant features It was a perfect example of how strong mathematical ideas become powerful engineering tools.\nWhy this matters # Modern ML is not only about deep learning frameworks.\nIt is about understanding:\nYour data Your features Your assumptions And that is where mathematics quietly does most of the heavy lifting.\nFinal thoughts # PCA is simple, old, and beautiful.\nAnd it still competes with much more complex techniques in real-world applications.\nThat is why I enjoy working in Machine Learning:\nIt is a field where theory and practice constantly meet.\nIf you have a strong mathematical background, ML is a natural place to apply it.\n","date":"14 August 2023","externalUrl":null,"permalink":"/posts/firstpost/","section":"","summary":"This is my first post on my site","title":"From Mathematics to Machine Learning: Why PCA Is Still One of My Favorite Tools","type":"posts"},{"content":"","date":"14 August 2023","externalUrl":null,"permalink":"/tags/space/","section":"Tags","summary":"","title":"Space","type":"tags"},{"content":"","date":"14 August 2023","externalUrl":null,"permalink":"/tags/","section":"Tags","summary":"","title":"Tags","type":"tags"},{"content":" SHORT BIO # I’m a Machine Learning Engineer with an MSc in Mathematics from TU Munich and experience building real-world ML systems in industry. My work focuses on Python-based machine learning, computer vision, and data-driven solutions, combining strong mathematical foundations with practical engineering. I’ve worked on image classification, clustering, and deploying models using AWS. I’m interested in applied ML roles, research-oriented projects, and collaborations at the intersection of mathematics and its applications.\nApart from that, I am a proficient violinist and a very good football player. I speak fluent English and German.\nEDUCATION # Technical University of Munich 2021 - 2024 M.Sc Mathematics I specialized in Algebra and Geometry. I also did course work in applied mathematics in topics such as Deep Learning, General Relativity and Optimization. My thesis was in the area of Quantum Information. Shiv Nadar University 2017 - 2021 B.Sc Mathematics I graduated with Distinction and my thesis concerned itself with Vector Bundles. Journal Articles # An Automated Prefabricated Facade Layout Definition for Residential Building Renovation\nKepa Iturralde, Samanti Das, Aravind Srinivasaragavan, Thomas Bock, Christoph Holst\nBuildings, 13(12), 2023.\nDOI: 10.3390/buildings13122981\nEnhanced Precision in Built Environment Measurement: Integrating AprilTags Detection with Machine Learning\nShengtao Tan, Aravind Srinivasaragavan, Kepa Iturralde, Christoph Holst\n, 1295–1298, 2024.\nDOI: 10.22260/ISARC2024/0167\nA Robust Pipeline for Facade Geometry Estimation Using AprilTags and Digital Imaging\nAravind Srinivasaragavan, Danya Liu, Kepa Iturralde, Christoph Holst\n, 1530–1535, 2025.\nDOI: 10.22260/ISARC2025/0199\nIntegrated Workflow for a Semi-Automated and Robotic Facade Renovation\nKepa Iturralde, Wenlan Shen, Renzo Bazan, Oscar Amo Grau, …, Aravind Srinivasaragavan, Samanti Das, Thomas Bock, Christoph Holst\nFrontiers in Built Environment, 11, 2025.\nDOI: 10.3389/fbuil.2025.1649278\n","externalUrl":null,"permalink":"/profile/","section":"","summary":"","title":"","type":"page"},{"content":"","externalUrl":null,"permalink":"/authors/","section":"Authors","summary":"","title":"Authors","type":"authors"},{"content":"","externalUrl":null,"permalink":"/categories/","section":"Categories","summary":"","title":"Categories","type":"categories"},{"content":"","externalUrl":null,"permalink":"/series/","section":"Series","summary":"","title":"Series","type":"series"}]