New opportunity

Algorithm Engineer (R&D)

KLA · Singapore

About this role

The Algorithm Engineer (R&D) will focus on developing image and signal processing algorithms for next-generation optical inspection and metrology systems within KLA's Global Products Group. Responsibilities include the full algorithm life cycle: modeling, proof-of-concept design, production software implementation, performance characterization, documentation, and user support. The role requires significant cross-functional collaboration for prototyping and testing. Requirements: Proven innovative track record and fundamental knowledge in algorithm development, specifically texture analysis, deep learning, image segmentation, feature extraction, classification, statistical data analysis, signal processing, and filter theory. Proficiency in C/C++, OOD/OOP, MATLAB, and/or Python is required. Preferred qualifications include experience in algorithm performance optimization, CUDA programming, parallel computing, or distributed computing. A Ph.D. or Master's degree in Electrical Engineering, Computer Science, or Physics is preferred. Candidates must be self-motivated with strong communication skills for working within a multi-disciplinary global team.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

Algorithm DevelopmentTexture AnalysisDeep LearningImage SegmentationFeature ExtractionClassificationStatistical Data AnalysisSignal ProcessingFilter TheoryAlgorithm Performance OptimizationCUDAParallel ComputingDistributed ComputingC++COODOOPMATLABPython

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

Full-Cycle Algorithm Development
The role demands ownership of the entire algorithm lifecycle, from initial mathematical modeling and proof-of-concept design to production-grade software implementation and performance characterization. You must demonstrate the ability to transition theoretical research into robust, deployable code.
Cross-Functional Collaboration
Because algorithms are central to optical inspection systems, you will spend significant time collaborating with multi-disciplinary teams. You must be prepared to communicate complex technical concepts to non-algorithm specialists and integrate feedback from hardware and software engineers during prototyping and testing.
Core Technical Proficiency
You must possess deep expertise in image and signal processing, specifically in texture analysis, deep learning, and image segmentation. Proficiency in C/C++ and OOD/OOP is essential for production implementation, while MATLAB or Python is required for rapid prototyping and data analysis.

Likely challenges to think through

Balancing Innovation with Production Constraints
A common challenge in R&D for semiconductor equipment is ensuring that advanced algorithms, such as deep learning models, meet strict real-time performance requirements. Prepare by discussing how you optimize code for latency and throughput, potentially using CUDA or parallel computing to meet production-level speed.
Navigating Multi-Disciplinary Complexity
The intersection of physics, optics, and software engineering creates a high-complexity environment. You may face difficulty when your algorithm's performance is limited by hardware constraints. Prepare by highlighting examples where you successfully aligned software solutions with physical system limitations through iterative testing.

Practice your answers

Technical practice

Image Segmentation and Feature Extraction

What to coverDescribe your approach to segmenting complex wafer surface images to isolate potential defects. Discuss how you select and validate features for classification, especially when dealing with high-noise environments or varying texture patterns.

Deep Learning for Defect Classification

What to coverExplain how you would design a deep learning model for classifying wafer defects given limited labeled data. Address how you handle class imbalance and ensure the model generalizes well to new, unseen defect types.

Algorithm Performance Optimization

What to coverDiscuss strategies for optimizing image processing algorithms for real-time inspection systems. Suggestion: Focus on your experience with C++ memory management, algorithmic complexity reduction, or leveraging hardware acceleration like CUDA.

Signal Processing and Filter Theory

What to coverHow do you apply filter theory to remove noise from optical inspection data while preserving critical defect information? Explain the trade-offs between spatial and frequency domain filtering in the context of metrology.

Parallel and Distributed Computing

What to coverDescribe a scenario where you implemented parallel or distributed computing to speed up a computationally intensive algorithm. Explain how you managed data synchronization and load balancing across multiple processing units.

C++ and Object-Oriented Design

What to coverHow do you ensure your production-level C++ code is maintainable and scalable for long-term R&D projects? Discuss your use of design patterns or OOD principles to facilitate modular algorithm development.

Statistical Data Analysis

What to coverHow do you use statistical methods to characterize the performance of an algorithm during the R&D phase? Explain how you determine if an improvement in accuracy is statistically significant versus noise.

Algorithm Life Cycle Management

What to coverWalk through your process for transitioning an algorithm from a MATLAB proof-of-concept to a production-ready C++ implementation. Highlight how you handle performance characterization and documentation during this transition.

Behavioural practice

Cross-Functional Collaboration

What to coverDescribe a time you had to collaborate with a non-algorithm team (e.g., hardware or systems engineering) to prototype a solution. How did you communicate technical constraints and ensure the final design met system-level requirements?

Handling Technical Ambiguity

What to coverProvide an example of a project where the requirements were initially unclear or the problem was highly complex. What steps did you take to define the scope and move toward a successful proof-of-concept?

Communicating Complex Concepts

What to coverTell me about a time you had to explain a complex technical finding or algorithm limitation to a stakeholder or team member with a different background. How did you ensure your message was understood and actionable?

Self-Motivation and Independent Research

What to coverDescribe a situation where you identified a technical challenge that required you to learn a new tool or methodology independently. How did you approach the learning process and apply it to your work?

Sharpen your knowledge

Topics to prepare

Image and Signal Processing Fundamentals
Review core concepts in filter theory, feature extraction, and texture analysis. Exercise: Implement a custom spatial filter in MATLAB or Python to detect specific patterns in noisy images, then document the performance trade-offs. This is essential for the role's focus on optical inspection and metrology systems.
Algorithm Life Cycle Management
Understand the transition from proof-of-concept modeling to production-grade software implementation. Exercise: Take a prototype algorithm you have developed and refactor it for modularity and performance, focusing on how you would document it for a cross-functional team. This directly addresses the requirement for managing the full algorithm life cycle.
Statistical Data Analysis
Refresh knowledge on statistical methods for performance characterization and defect detection. Exercise: Analyze a dataset of inspection results to identify outliers or process non-uniformities using statistical techniques. This is relevant for the role's responsibility in monitoring process tools and identifying yield excursions.

Skills to revise

C++ and Object-Oriented Programming
Practice implementing complex algorithms using C++ with a focus on OOD/OOP principles. Exercise: Create a class hierarchy for an image processing pipeline that allows for interchangeable filter modules. This is a core requirement for production software implementation at KLA.
Algorithm Performance Optimization
Focus on techniques to improve execution speed for high-throughput inspection. Exercise: Profile a computationally intensive algorithm and identify bottlenecks, then apply optimization techniques such as vectorization or memory management. This is a preferred qualification that directly impacts system throughput.
CUDA and Parallel Computing
Study parallel programming models to accelerate image processing tasks. Exercise: Parallelize a simple image segmentation or convolution operation using CUDA kernels. This is highly relevant for the preferred qualification in parallel and distributed computing for next-generation inspection systems.

Your preparation checklist

Prepare a technical project portfolio
Select 2-3 past projects that demonstrate the full algorithm lifecycle, from initial modeling and proof-of-concept in MATLAB or Python to production-level implementation in C++. Focus on projects involving image segmentation, feature extraction, or classification.
Review core signal and image processing theory
Refresh your knowledge on filter theory, statistical data analysis, and texture analysis. Be prepared to explain how these concepts are applied to noise reduction or defect detection in high-precision optical systems.
Demonstrate performance optimization skills
Prepare to discuss how you have optimized algorithms for speed and efficiency. If you have experience with CUDA, parallel computing, or distributed computing, create a brief summary of a specific bottleneck you resolved and the resulting performance gains.
Practice cross-functional communication
Prepare examples of how you have collaborated with non-algorithm teams (such as hardware or systems engineers) to prototype, test, and integrate your software. Emphasize your ability to translate complex technical requirements into actionable development steps.
Formulate questions for the interviewers
Prepare thoughtful questions about the current challenges in optical inspection, the typical size and structure of the multi-disciplinary teams you would work with, and how the team balances R&D innovation with production-grade software stability.

Availability last checked: 18 Oct 2026. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…