Pawan kumar

- Lead development of multimodal AI, NLP, and LLM solutions, boosting user satisfaction by 25% with RAG-powered applications. - Develop scalable, secure production-grade ML APIs for seamless client integration. - Optimize MLOps pipelines and cloud infrastructure, reducing costs by 40% and accelerating delivery. - Drive innovation in AI with advanced techniques like synthetic data and multi-agent architectures. - Mentor teams, fostering a culture of collaboration and responsible AI practices.

1. Led migration of applications and projects to GCP, optimizing infrastructure and enhancing scalability. 2. Forecasted and reduced compute costs for ML lifecycle. 3. Developed and implemented CI/CD and ML pipelines using MLflow, Kubeflow, and TFX. 4. Ensured scalability, security, and governance of ML infrastructure through maintenance and optimization. 5. Productionalized Generative AI solutions, including RagPowered Engines/Bots and a Prompt Engineering SaaS platform, focusing on AI and data governance.

Worked on the EU-funded Braine project at Leibniz Universität Hannover, focusing on scheduling strategies to improve cloud computing decision-making using ML. **Key Responsibilities:** 1. Researched supervised and reinforcement learning algorithms. 2. Designed products using SOTA methods for production migration. 3. Wrote and published research papers. 4. Engaged in Agile development. **Version Control and CI/CD:** GitLab, K8s, Docker **More details:** [Braine Project](https://www.braine-project.eu/)

1. Managed installation, configuration, administration, and automation of WebSphere Application Server and Oracle WebLogic Server. 2. Set up security permissions in WebSphere environments to authenticate against LDAP servers and secure consoles using SSL. 3. Troubleshot JEE runtime environment issues by analyzing server logs, diagnosing thread-dumps, and heap-dumps using MAT, IBM heap analyzer, and Oracle Visual JVM.