Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m Sri Navya K, a senior AI/ML and data engineer with 7+ years of experience building production-grade machine learning and NLP systems. My work spans end-to-end ML pipelines, transformer-based solutions for entity extraction and semantic search, and scalable real-time inference services using Python, SQL, Spark, Kafka, and modern ML frameworks like PyTorch and TensorFlow. I enjoy turning unstructured text and streaming data into reliable, measurable outcomes—whether that’s fine-tuning BERT/RoBERTa models for NER, creating recommendation systems with semantic embeddings, or integrating GenAI (AWS Bedrock and Azure OpenAI) for summarization and information validation. I also focus heavily on MLOps: CI/CD, monitoring, drift detection, model versioning, and automated retraining to keep models dependable in production.…I’m Sri Navya K, a senior AI/ML and data engineer with 7+ years of experience building production-grade machine learning and NLP systems. My work spans end-to-end ML pipelines, transformer-based solutions for entity extraction and semantic search, and scalable real-time inference services using Python, SQL, Spark, Kafka, and modern ML frameworks like PyTorch and TensorFlow. I enjoy turning unstructured text and streaming data into reliable, measurable outcomes—whether that’s fine-tuning BERT/RoBERTa models for NER, creating recommendation systems with semantic embeddings, or integrating GenAI (AWS Bedrock and Azure OpenAI) for summarization and information validation. I also focus heavily on MLOps: CI/CD, monitoring, drift detection, model versioning, and automated retraining to keep models dependable in production.

Sri Navya K

AI Engineer, Data Engineer, Data Scientist, +2





I’m Sri Navya K, a senior AI/ML and data engineer with 7+ years of experience building production-grade machine learning and NLP systems. My work spans end-to-end ML pipelines, transformer-based solutions for entity extraction and semantic search, and scalable real-time inference services using Python, SQL, Spark, Kafka, and modern ML frameworks like PyTorch and TensorFlow. I enjoy turning unstructured text and streaming data into reliable, measurable outcomes—whether that’s fine-tuning BERT/RoBERTa models for NER, creating recommendation systems with semantic embeddings, or integrating GenAI (AWS Bedrock and Azure OpenAI) for summarization and information validation. I also focus heavily on MLOps: CI/CD, monitoring, drift detection, model versioning, and automated retraining to keep models dependable in production.…I’m Sri Navya K, a senior AI/ML and data engineer with 7+ years of experience building production-grade machine learning and NLP systems. My work spans end-to-end ML pipelines, transformer-based solutions for entity extraction and semantic search, and scalable real-time inference services using Python, SQL, Spark, Kafka, and modern ML frameworks like PyTorch and TensorFlow. I enjoy turning unstructured text and streaming data into reliable, measurable outcomes—whether that’s fine-tuning BERT/RoBERTa models for NER, creating recommendation systems with semantic embeddings, or integrating GenAI (AWS Bedrock and Azure OpenAI) for summarization and information validation. I also focus heavily on MLOps: CI/CD, monitoring, drift detection, model versioning, and automated retraining to keep models dependable in production.

Available to hire

I’m Sri Navya K, a senior AI/ML and data engineer with 7+ years of experience building production-grade machine learning and NLP systems. My work spans end-to-end ML pipelines, transformer-based solutions for entity extraction and semantic search, and scalable real-time inference services using Python, SQL, Spark, Kafka, and modern ML frameworks like PyTorch and TensorFlow.

I enjoy turning unstructured text and streaming data into reliable, measurable outcomes—whether that’s fine-tuning BERT/RoBERTa models for NER, creating recommendation systems with semantic embeddings, or integrating GenAI (AWS Bedrock and Azure OpenAI) for summarization and information validation. I also focus heavily on MLOps: CI/CD, monitoring, drift detection, model versioning, and automated retraining to keep models dependable in production.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

Senior AI/ML Engineer at State of VA

October 1, 2024 - Present

Built production-ready Named Entity Recognition (NER) solutions using spaCy and transformer models (BERT/RoBERTa) to extract structured fields from unstructured government documents. Developed end-to-end ML pipelines for ingestion, preprocessing, feature engineering, training, evaluation, deployment, and monitoring. Implemented semantic search and auxiliary classification tasks using fine-tuned transformer models and embedding generation. Served models via REST APIs (FastAPI/Flask) and automated deployment with Docker, Kubernetes, Jenkins, Git, and MLflow. Integrated AWS Bedrock and Azure OpenAI to add a supplementary layer for summarization and validation alongside core NER models. Added production safeguards including monitoring, drift detection, explainability (SHAP), and automated retraining, collaborating in an Agile environment and mentoring junior engineers.

AI/ML Engineer at GM — Warren

May 1, 2023 - September 1, 2024

Designed and built an NLP-based recommendation engine to match candidates to job openings using semantic similarity and embedding techniques. Created ML pipelines covering data preparation, feature engineering, model training, evaluation, deployment, and monitoring. Generated embeddings using Sentence Transformers, BERT, TF-IDF, and Word2Vec; built supporting predictive models (e.g., XGBoost, Random Forest, LightGBM, Logistic Regression) and combined them with ranking logic using cosine similarity and model scores. Improved performance with feature selection, cross-validation, and hyperparameter tuning, and used SHAP/LIME for interpretability. Deployed containerized services using FastAPI, Docker, and Kubernetes with CI/CD automation (MLflow, Git, Jenkins), and implemented monitoring for accuracy and data drift with retraining triggers.

AI/ML Engineer at AT&T

September 1, 2021 - March 1, 2023

Built ETL and streaming pipelines for telecom data using Python, SQL, Apache Spark, and Kafka to support downstream ML workloads. Developed feature engineering workflows for predictive analytics and customer behavior modeling. Trained classical ML models (Random Forest, XGBoost, LightGBM, Logistic Regression) and implemented deep learning solutions with TensorFlow and PyTorch for time-series forecasting and anomaly detection. Automated workflows using Apache Airflow and MLflow with CI/CD practices. Deployed training and inference workflows using AWS services (SageMaker, S3, Glue, Redshift) and supported large-scale processing with Databricks and Snowflake. Conducted EDA, statistical analysis, feature selection, and built business-facing dashboards using Matplotlib and Plotly.

AI/ML Engineer at Quadrant Resources

November 1, 2019 - August 1, 2021

Developed batch and real-time ETL pipelines for healthcare and recruitment datasets, extending them with predictive and NLP solutions for analytics and document classification. Built structured-data ML models with Python and scikit-learn/SQL and created preprocessing and feature engineering workflows for both structured and unstructured data. Trained models such as Logistic Regression, Random Forest, and XGBoost ensembles. Implemented BERT-based NLP pipelines for medical document classification and text analysis and developed deep learning models with TensorFlow and PyTorch for sequential data. Used dbt to create and maintain transformation models feeding consistent marts, and deployed services with FastAPI and Docker via CI/CD. Produced performance and business insight reports with Matplotlib/Plotly while collaborating with stakeholders in an Agile setting.