Akarsh Reddy A.

About Akarsh Reddy A.

I am a Data Scientist with over 3 years of experience working with real-time telemetry from millions of servers and components to build predictive models, detect anomalies, and monitor operations. In my current projects, I develop and deploy AI-driven chatbots that give operators instant access to troubleshooting, diagnostics, and documentation, while applying analytics to optimize supply chain operations and improve efficiency. Using Python, PySpark, SQL, Snowflake, and AWS, I have built scalable solutions that improved defect prediction accuracy by 20%, reduced testing errors by 15%, and accelerated decision-making by 30%.My experience in data centers, marketing analytics, and supply chain optimization gives me a unique perspective for finance use cases such as risk modeling, forecasting, fraud detection, pricing optimization, and operational intelligence. I enjoy turning complex data into clear, actionable insights and delivering AI/ML solutions that are reliable, scalable, and impactful.I enjoy collaborating with cross-functional teams and sharing ideas. I’m always happy to connect with professionals working on AI/ML, data-driven solutions, or innovative analytics projects.

Experience

1. Data Scientist

   Jabil

   Mar 2025 — Present · Florence, KY, US

   Analyzed centralized data with over 5M server, component, and telemetry using Python (Pandas, NumPy), PySpark, SQL, and Snowflake, identifying recurring failure signatures and performance bottlenecks that reduced testing errors by 15% and improved diagnostic throughput across production lines.•Collaborated with R&D, IT Infrastructure, Reliability and Test Engineering, and Business Operations to collect, cleanse, and interpret large-scale system data from distributed test benches, enabling data-driven upgrades to server diagnostics, hardware validation, and predictive maintenance processes.•Built and deployed regression, classification, and clustering models (XGBoost, LightGBM, SVM, Logistic Regression, K-Means) on AWS Sagemaker, Azure ML, and GCP Vertex AI, improving defect prediction accuracy by 20% and accelerating component triage and failure categorization.•Architected real-time monitoring dashboards using Power BI, Grafana, and Python Dash/Plotly, integrating cloud-based metrics from AWS CloudWatch, Azure Monitor, and Prometheus, reducing data-to-decision latency by 30% for engineering and operations teams.•Performed extensive statistical modeling, hypothesis testing, anomaly detection, and cross-validation of ML pipelines, ensuring high-confidence predictions and enabling 25% faster test-cycle adjustments in automated diagnostic systems.•Supported end-to-end ML and analytics operations, including ETL pipeline development (Airflow, dbt), model training, Docker/Kubernetes deployment, CI/CD (GitHub Actions, Jenkins), model monitoring, drift detection, and re-training, improving model reliability by 18%.•Implemented NLP, LLM-based log analysis, and Generative AI techniques to extract insights from multi-terabyte unstructured logs, enabling early anomaly detection and improving root-cause identification accuracy by 22% across hardware test environments.

Education

• Gokaraju Rangaraju Institute of Engineering and Technology

  Bachelor of Technology - BTech, Information Technology

• University of Cincinnati

  Master of Science - MS, Information Technology