Si-Ware Systems is a global leader in semiconductor and spectroscopy solutions. Our innovative devices and software enable material analysis across many industries.
At Si-Ware, we foster a culture of innovation, collaboration, and continuous learning, empowering our people to push the boundaries of technology.
As a Machine Learning Engineer at Si-Ware, you will design, implement, and deploy applied ML solutions that power real-world spectroscopy devices and emerging physical AI systems.
You will work at the intersection of machine learning, software engineering, and intelligent hardware integration.
initiatives.
**Responsibilities:
Machine Learning & Modeling**
Perform data cleaning, preprocessing, and transformation for training and evaluation.
Contribute to the development, training, and improvement of machine learning models.
Design and execute structured experiments using appropriate validation strategies (cross-validation, hold-out testing, statistical comparison).
Evaluate models using appropriate performance metrics (Accuracy, Precision, Recall, F1, RMSE, etc.).
Optimize models and inference pipelines for performance, memory efficiency, and real-time constraints when required.
Ensure reproducibility, traceability, and proper validation of models within regulated or industrial environments.
Support chemometrics-related workflows, including spectral preprocessing, feature extraction, multivariate modeling, and validation for spectroscopy-based applications.
Production & System Integration
- Design and implement production-ready ML pipelines integrated with software applications and hardware systems.
- Ensure models are maintainable, version-controlled, and deployable within real-world products.
- Contribute to the design and evolution of ML system architecture and reusable pipeline components.
- Develop unit and integration tests for ML components to ensure reliability.
- Document ML modules and interfaces to support long-term maintainability.
- Work closely with software and firmware teams to integrate ML models into desktop applications, services, and embedded workflows.
Tools, Innovation & Growth
- Participate in the development of internal and customer-facing ML-driven tools and automation utilities.
- Stay updated with the latest machine learning research, tools, and industry practices.
Contribute to exploratory and applied ML solutions in emerging physical AI systems, including robotics and sensor-driven intelligent platforms.
Bachelor’s degree in Computer Science, Engineering, Mathematics, or a related field.
1 - 4 years of relevant industry or applied ML experience preferred.
Strong focus on applied machine learning and engineering implementation.
Hands-on experience with Python and its ML ecosystem (NumPy, Pandas, Matplotlib, Scikit-Learn).
Familiarity with at least one deep learning framework (PyTorch or TensorFlow).
Solid understanding of software engineering principles (modular design, version control, testing, debugging).
Experience structuring modular Python codebases.
Familiarity with packaging, model serialization, and reproducible pipelines.
Experience working within larger software systems (not only notebooks).
Strong analytical and problem-solving skills.
Good communication skills and ability to work in a collaborative team environment.
Proficiency in English (reading and writing).
Nice to Have:
- Basic understanding of chemometrics concepts (multivariate analysis, regression/classification, spectral data handling).
- Exposure to spectroscopy data (NIR, IR) and common preprocessing techniques (normalization, smoothing, baseline correction).
- Experience contributing to ML tools, internal platforms, or data analysis software.
- Knowledge of MLOps practices (Git, CI/CD, Docker).
- Experience with cloud platforms (AWS, GCP, Azure).
- Experience with robotics frameworks (ROS).
- Experience with sensor fusion.
- Experience with real-time ML inference.
- Basic understanding of control systems.
- Exposure to data visualization or BI tools.
- Contributions to Kaggle competitions, research projects, or open-source initiatives.