Applied AI Engineer at PTC

Improved engineering support efficiency by 28% through fine-tuning internal LLMs with QLoRA and PEFT, integrating with LangChain and RAG for precise technical answers. Developed real-time monitoring dashboards using Power BI, Prometheus, and Grafana to maintain deployed model uptime and performance. Automated AR content creation with Stable Diffusion and OpenCV, reducing manual effort by 40%. Collaborated with product teams and domain experts to align AI solutions with business needs, ensured responsible AI adoption with explainability tools, and accelerated model deployment by 60% using Kubeflow, MLflow, and AWS SageMaker.

Junior AI/ML Engineer – Automotive Data & Diagnostics at ARAPL

Developed predictive maintenance models using XGBoost and LightGBM on connected vehicle sensor data, reducing component failures by 25%. Built real-time anomaly detection systems processing 100k+ telemetry signals daily using PySpark and Kafka, cutting diagnostic error reporting time by 30%. Implemented data ingestion pipelines with Apache Spark and Dask integrating CAN bus and OBD-II data. Fine-tuned computer vision models with OpenCV to detect mechanical wear, streamlining manual inspections. Created interactive dashboards in Power BI and Tableau to visualize vehicle health metrics and fault codes, collaborating with automotive experts to ensure practical AI applicability.

Applied AI Engineer at PTC

Improved engineering support efficiency by 28% by fine-tuning internal LLMs with QLoRA and PEFT techniques, integrating them into LangChain and RAG pipelines for precise answers from extensive technical documentation. Developed real-time dashboards using Power BI, Prometheus, and Grafana to monitor models for uptime and performance. Automated AR content creation by 40% through deploying computer vision pipelines with Stable Diffusion and OpenCV. Collaborated with product managers and domain experts to ensure AI solutions align with user needs. Drove responsible AI by integrating explainability tools like SHAP and LIME. Accelerated model deployment by 60% by designing production-grade MLOps frameworks using Kubeflow, MLflow, and AWS SageMaker.

Junior AI/ML Engineer – Automotive Data & Diagnostics at ARAPL

Developed predictive maintenance models using XGBoost and LightGBM on connected vehicle sensor data, reducing failures by 25%. Designed and deployed real-time anomaly detection pipelines with PySpark and Kafka, processing over 100K telemetry records daily, reducing diagnostic error reporting time by 30%. Built data ingestion workflows integrating CAN bus, OBD-II logs, and third-party APIs with Apache Spark and Dask. Fine-tuned OpenCV-powered image models for mechanical wear detection from workshop camera feeds, streamlining inspections. Created interactive diagnostic dashboards with Power BI and Tableau for vehicle health KPIs. Collaborated with domain experts to ensure practical AI application in service workflows.

Associate Data Scientist at ARAPL

Built machine learning models with Scikit-learn and XGBoost to predict vehicle part failures, improving maintenance scheduling by 22%. Automated ETL pipelines using Python, Pandas, and SQL to reduce data preparation time by 40%. Collaborated on data labeling and cleaning for CAN bus and OBD-II logs. Designed Power BI dashboards presenting fault trends and sensor anomalies. Applied SHAP to enhance the interpretability of model predictions. Participated in Agile sprints gaining full ML development lifecycle exposure in automotive diagnostics.

Applied AI Engineer at PTC

Led production-grade AI initiatives: fine-tuned internal LLMs with QLoRA and PEFT, integrated LangChain-based RAG pipelines powered by FAISS for vector search, improving engineering support accuracy and reducing response time by 28%. Collaborated with cross-functional teams to deploy scalable ML solutions across auto and industrial contexts.

Data Scientist (Automotive Data & Diagnostics) at ARAPL Labs

Developed predictive maintenance and diagnostics models on automotive sensor data, reducing downtime and improving fleet uptime. Applied Python and PySpark for data processing; collaborated with domain experts to deliver robust ML solutions.