About the Role 

We are seeking a Principal Machine Learning Engineer – Model Efficiency & Optimization to serve as the technical anchor for ABBYY’s model optimization strategy. 

This is a senior individual contributor role for a deep domain expert who will define how ABBYY builds efficient, high-performing, production-ready models for document AI at scale. You will set technical direction from research exploration through production deployment, combining strong theoretical expertise with hands-on implementation. 

Key Responsibilities 

Research Direction & Technical Strategy 

• Own the end-to-end technical direction for model efficiency and optimization, from research agenda to production deployment 

• Define approaches for building efficient, production-ready models optimized for document AI use cases 

• Establish frameworks for evaluating quality vs. efficiency trade-offs (accuracy, latency, memory footprint) 

• Set standards for what constitutes a successful optimized model across document understanding benchmarks 

• Evaluate and adopt emerging techniques in model optimization and compression 

• Influence modeling strategy across teams by integrating efficiency-first thinking into model development 

 Hands-on Implementation & Experimentation 

• Lead design and implementation of optimization pipelines, training objectives, and compression techniques 

• Run large-scale experiments and analyze training dynamics, instabilities, and capability gaps 

• Develop novel optimization approaches tailored to document understanding tasks, including layout and multimodal challenges 

• Diagnose and resolve failure modes such as quality degradation and generalization gaps 

• Prototype and validate new techniques before scaling to production training runs 

 Cross-Functional Collaboration 

• Partner with the Document AI Data team to define training data requirements for optimized models 

• Collaborate with Platform teams on distributed training infrastructure, experiment tracking, and compute strategy 

• Work closely with Modeling teams to ensure optimized models meet quality and performance standards 

• Communicate technical trade-offs, findings, and recommendations clearly to engineering, product, and leadership stakeholders 

Qualifications

Education & Experience 

• MS or PhD in Computer Science, Engineering, Mathematics, or related field (PhD preferred) 

• 10+ years of experience in Machine Learning / AI, with focus on:  

• Model optimization 

• Efficient deep learning 

• Large-scale model deployment 

• Demonstrated track record of contributions to model efficiency (e.g., publications, patents, or industry impact) 

• Proven experience optimizing large-scale language and/or vision models for production 

• Deep understanding of trade-offs between model quality, size, and inference performance 

 Technical Expertise 

• Deep expertise in model optimization and compression techniques (e.g., quantization, pruning) 

• Strong knowledge of efficient deep learning methodologies 

• Expertise in Vision-Language Models (VLMs) and multimodal optimization challenges 

• Strong programming skills in Python and deep proficiency with PyTorch or similar frameworks 

• Experience with distributed training systems and large-scale experimentation workflows 

• Strong evaluation methodology for optimized models, including benchmarking, efficiency profiling, and regression analysis 

 Leadership & Communication 

• Recognized technical authority in model optimization or efficient AI systems 

• Proven ability to influence technical direction without formal authority 

• Strong track record of driving applied research → production impact 

• Excellent communication skills, with the ability to clearly articulate complex trade-offs 

• Collaborative mindset with the ability to align cross-functional teams 

Here are some of our local benefits: 

• Comprehensive medical, accidental, and life insurance 

• Weekly wellness sessions to support your physical and mental well-being 

• A generous paid time off policy