Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am Saiyam Jain, a Machine Learning Engineer based in Karlsruhe, Germany, specializing in end-to-end AI solutions and data-driven decision-making. With strong expertise in Computer Vision, NLP, Generative AI, MLOps, and cloud-native ML systems such as AWS, Azure, and GCP, I excel in building, optimizing, deploying, and maintaining production-level AI models. I am passionate about leveraging AI-driven solutions to analyze complex data and solve real-world problems. Over the years, I have collaborated with academic and industry teams, supervised students, and published research on adversarial attacks, quantization, and hardware-aware machine learning, continually advancing my skills in AI and deep learning domains.…I am Saiyam Jain, a Machine Learning Engineer based in Karlsruhe, Germany, specializing in end-to-end AI solutions and data-driven decision-making. With strong expertise in Computer Vision, NLP, Generative AI, MLOps, and cloud-native ML systems such as AWS, Azure, and GCP, I excel in building, optimizing, deploying, and maintaining production-level AI models. I am passionate about leveraging AI-driven solutions to analyze complex data and solve real-world problems. Over the years, I have collaborated with academic and industry teams, supervised students, and published research on adversarial attacks, quantization, and hardware-aware machine learning, continually advancing my skills in AI and deep learning domains.

Saiyam Jain





I am Saiyam Jain, a Machine Learning Engineer based in Karlsruhe, Germany, specializing in end-to-end AI solutions and data-driven decision-making. With strong expertise in Computer Vision, NLP, Generative AI, MLOps, and cloud-native ML systems such as AWS, Azure, and GCP, I excel in building, optimizing, deploying, and maintaining production-level AI models. I am passionate about leveraging AI-driven solutions to analyze complex data and solve real-world problems. Over the years, I have collaborated with academic and industry teams, supervised students, and published research on adversarial attacks, quantization, and hardware-aware machine learning, continually advancing my skills in AI and deep learning domains.…I am Saiyam Jain, a Machine Learning Engineer based in Karlsruhe, Germany, specializing in end-to-end AI solutions and data-driven decision-making. With strong expertise in Computer Vision, NLP, Generative AI, MLOps, and cloud-native ML systems such as AWS, Azure, and GCP, I excel in building, optimizing, deploying, and maintaining production-level AI models. I am passionate about leveraging AI-driven solutions to analyze complex data and solve real-world problems. Over the years, I have collaborated with academic and industry teams, supervised students, and published research on adversarial attacks, quantization, and hardware-aware machine learning, continually advancing my skills in AI and deep learning domains.

Available to hire

I am Saiyam Jain, a Machine Learning Engineer based in Karlsruhe, Germany, specializing in end-to-end AI solutions and data-driven decision-making. With strong expertise in Computer Vision, NLP, Generative AI, MLOps, and cloud-native ML systems such as AWS, Azure, and GCP, I excel in building, optimizing, deploying, and maintaining production-level AI models.

I am passionate about leveraging AI-driven solutions to analyze complex data and solve real-world problems. Over the years, I have collaborated with academic and industry teams, supervised students, and published research on adversarial attacks, quantization, and hardware-aware machine learning, continually advancing my skills in AI and deep learning domains.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Intermediate

Language

English

Advanced

German

Intermediate

Hindi

Fluent

Work Experience

Deep Learning Researcher at Karlsruhe Institute of Technology

May 1, 2023 - Present

Trained GAN and diffusion models for fracture path prediction in 3D brittle-anisotropic materials, achieving 96% structural similarity in predicted paths. Led collaboration with research teams and supervised master’s students on integration of physics-informed neural networks. Fine-tuned and deployed large language models such as GPT and LLaMa as scalable query-based search tools via FastAPI, implementing retrieval-augmented generation (RAG) to improve internal search efficiency and user experience on the institute’s website.

Computer Vision Engineer at Center for Advanced Systems Understanding

January 31, 2023 - August 7, 2025

Developed defense pipelines against adversarial attacks on ML image denoisers which improved medical image segmentation F1 score by 20%. Optimized high resolution microscopic imaging inference latency using CUDA and TensorFlow. Published research work on adversarial attacks and defenses for ML image denoisers.

Deep Learning Engineer at Fraunhofer IPMS

August 31, 2022 - August 7, 2025

Designed quantized neural networks for ASICs, cutting memory footprint of CNNs like ResNet-20 by 73%. Applied mixed-precision quantization and model compression for hardware-efficient machine learning prototyping. Delivered proof-of-concept models for real-time edge device inference. Published work on hardware-aware mixed quantization.

Artificial Intelligence Engineer at National Centre for Tumor Diseases

July 31, 2021 - August 7, 2025

Designed evaluation framework for Explainable AI (XAI) using surgical imaging datasets. Benchmarked multiple XAI methods on pretrained AlexNet models and developed lightweight visualization interfaces for interpretability metrics.

Data Scientist at Helmholtz Zentrum Dresden Rossendorf

March 31, 2021 - August 7, 2025

Developed neural network surrogate models for 3D current deposition based on LWFA simulation data and point cloud processing to improve deposition approximation for particle-in-cell simulations. Constructed interactive dashboards for hyperparameter and model metric evaluations facilitating performance comparisons.

Deep Learning Researcher at Karlsruhe Institute of Technology

May 1, 2023 - Present

Trained GAN and diffusion models for fracture path prediction in 3D brittle-anisotropic materials achieving 96% structural similarity of predicted paths. Led collaboration with research teams and supervised master's students to integrate Physics-Informed Neural Networks (PINNs). Fine-tuned and deployed large language models (GPT, LLaMa) as scalable query-based search tools using FastAPI, and implemented Retrieval-Augmented Generation (RAG) to improve internal search efficiency and user experience on the institute’s website.

Computer Vision Engineer at Center for Advanced Systems Understanding

January 31, 2023 - August 7, 2025

Developed a defense pipeline against adversarial attacks on machine learning image denoisers, improving medical image segmentation by 20% in F1 score. Optimized high-resolution microscopic imaging using CUDA and TensorFlow, reducing inference latency. Published work on adversarial attacks and defenses for ML image denoisers.

Deep Learning Engineer at Fraunhofer IPMS

August 31, 2022 - August 7, 2025

Designed quantized neural networks for ASICs, reducing CNN memory footprint by 73% with quantized ResNet-20. Applied mixed-precision quantization and model compression for hardware-efficient machine learning prototyping. Delivered proof-of-concept models for real-time inference on edge devices. Published work on hardware-aware mixed quantization.

Artificial Intelligence Engineer at National Centre for Tumor Diseases

July 31, 2021 - August 7, 2025

Designed an evaluation framework for Explainable AI (XAI) using surgical imaging datasets. Benchmarked multiple XAI methods on a pre-trained AlexNet model and developed lightweight interfaces to visualize interpretability metrics.

Data Scientist at Helmholtz Zentrum Dresden Rossendorf

March 31, 2021 - August 7, 2025

Developed a neural network-based surrogate model for 3D current deposition using LWFA simulation data and point cloud processing, improving deposition approximation for particle-in-cell simulation. Built interactive dashboards to evaluate hyperparameters and model metrics, enabling visualization and performance comparison.