Senior Machine Learning Engineer - Ai-assisted Data Annotation

ABBYY ABBYY · Enterprise · India · R&D (Engineering)

This role focuses on designing, building, and owning AI-assisted data annotation pipelines using LLMs and vision-language models to generate high-quality training data at scale for enterprise customers. It involves developing annotation strategies, quality validation systems, and optimizing inference pipelines.

What you'd actually do

  1. Design and implement AI-powered annotation pipelines using large models to generate ground truth labels at scale
  2. Build and optimize large-scale inference pipelines for processing millions of documents
  3. Create evaluation frameworks to benchmark automated annotations against human-labeled data
  4. Own the automated annotation track end-to-end, from architecture through production monitoring
  5. Implement monitoring and alerting for quality degradation and system failures

Skills

Required

  • Python
  • PyTorch
  • LLMs
  • VLMs
  • Data Annotation
  • Evaluation Design
  • Quality Measurement
  • Large-scale inference pipelines
  • Model serving systems
  • Human-in-the-loop systems

Nice to have

  • MS or PhD in Computer Science, Engineering, Mathematics, or related field
  • Document understanding tasks (classification, extraction, layout analysis, semantic parsing)
  • Label quality metrics
  • Confidence scoring
  • Agreement analysis

What the JD emphasized

  • 5+ years of experience in Machine Learning / AI
  • Large Language Models (LLMs)
  • Vision-Language Models (VLMs)
  • Data annotation or labeling systems
  • Demonstrated success using large AI models to automate annotation at production scale
  • Strong background in evaluation design and quality measurement
  • Deep expertise in LLMs and VLMs, including prompting, instruction tuning, and output evaluation
  • Strong understanding of document understanding tasks (classification, extraction, layout analysis, semantic parsing)
  • Experience designing label quality metrics, confidence scoring, and agreement analysis
  • Experience with large-scale inference pipelines and model serving systems
  • Familiarity with human-in-the-loop annotation systems and automation trade-offs

Other signals

  • Leveraging LLMs and vision-language models to generate high-quality training data at scale
  • Design and build AI-assisted annotation pipelines
  • Build and optimize large-scale inference pipelines for processing millions of documents