MARCUS JOHNSON

About

ML/NLP engineer focused on production-grade data and search/recommendation systems: built an end-to-end pipeline that connects unstructured customer feedback with product data using TF-IDF/BERT, Spark, and AWS (SageMaker/S3), orchestrated with Airflow and monitored for drift. Also has hands-on experience with entity resolution at scale and improving search relevance via BERT embeddings, FAISS vector search, and domain fine-tuning validated with precision@k and A/B testing.

Experience

Education

Key Strengths

• Built scalable NLP recommendation pipeline linking unstructured feedback to product data
• Production ML workflow orchestration with automated retraining/data refresh (Airflow)
• Model/embedding versioning and monitoring with drift detection
• Spark-based distributed processing for scale
• Entity resolution across multi-source customer records using fuzzy matching + rules
• Validation-driven approach (sampling, precision/recall, analyst feedback loops)
• Vector search implementation with BERT embeddings + FAISS to improve retrieval despite wording differences
• Domain fine-tuning with clear relevance metrics (precision@k) and hold-out testing
• Pragmatic prototype-to-production decisioning using baseline comparisons and latency targets
• Stakeholder alignment via dashboards and gradual rollout with A/B testing