Pavan Kumar Adapala

- Designed and maintained end-to-end ML pipelines using MLflow and DVC for experiment tracking, model versioning, and reproducible training workflows across multiple projects. - Automated model deployment and retraining workflows using Jenkins CI/CD pipelines, improving time-to-production for new model versions by 40%. Also Containerized ML workloads with Docker and orchestrated deployments on Kubernetes, and delivered scalable and consistent inference services.

- Developed and evaluated ML-based load forecasting models (batch and one-step-ahead) using IIoT time-series data from automotive paint shop HVAC systems, achieving 95% forecasting accuracy while accounting for weather-related uncertainties - Enhanced Predictive Energy Management by enabling proactive load shifting to reduce energy waste and operational costs, operationalizing models under real-world constraints of critical energy infrastructure in close collaboration with data scientists and domain experts

- Designed and optimized Jenkins CI/CD pipelines to enhance the development process of parking systems (ADAS) in a distributed, hardware–software integrated environment. Implemented data-driven automation strategies using Python and REST APIs to auto-email test summaries (unit, smoke, integration), reducing debugging time by 30%. - Built a centralized Grafana dashboard leveraging PostgreSQL queries to monitor CI/CD build KPIs across multiple projects, enabling data-driven release planning.

- Architected and managed cloud infrastructure on AWS using Terraform and Ansible, enforcing infrastructure-as-code practices that reduced provisioning time by 35% and ensured environment consistency across dev, staging, and production. Built and maintained Jenkins-based CI/CD pipelines for microservices applications, and Containerized applications with Docker and managed orchestration via Kubernetes, enabling zero-downtime deployments and improving release quality and deployment frequency.

- Contributed to cloud migration Proof-of-Concept (PoC) project, reducing infrastructure cost by 25% through implementation of scalable (Auto Scaling), production-ready AWS infrastructure. - Established GitHub repository standards including branch protection rules, mandatory CI checks before merges, and optimized branching strategy, reducing integration failures and improving code quality. Implemented CI/CD pipelines using Jenkins Master-Slave architecture to automate deployments.