Hire Ram D. - Machine Learning Engineer

At Pixis, I led the design and deployment of production-grade generative AI systems powering creative automation for product advertising. My work focused on building diffusion-based pipelines for ad content generation, enabling scalable, high-quality asset creation across multiple product lines. I architected and deployed these pipelines through FastAPI microservices, creating a modular AI backend that integrated seamlessly with client-facing applications. To ensure scalability and low-latency inference, I deployed models on platforms such as Replicate and Baseten, while also developing a Model Control Protocol (MCP) layer to orchestrate services, manage workloads, and optimize resource utilization. In addition to system design, I established and maintained CI/CD pipelines (GitHub Actions) to automate testing, deployment, and monitoring, which significantly improved release velocity and reliability. I collaborated closely with product and infra teams to align AI capabilities with business needs, driving adoption of generative pipelines in production. This role allowed me to contribute across the entire ML lifecycle — from research and prototyping to backend engineering, deployment, and monitoring — while mentoring junior engineers and ensuring engineering best practices.

* Develop and deploy cutting-edge computer vision solutions, including segmentation and pose detection algorithms, transforming theoretical concepts into practical applications. * Design and customize diffusion models pipelines tailored to specific business needs, ensuring seamless integration and optimal performance. * Fine-tune diffusion models with precision, aligning them intricately with evolving business requirements and technological advancements. * Stay at the forefront of the rapidly advancing GenAI landscape, with a particular focus on image generation techniques. * Maintain meticulous documentation, recording experiments, model architectures, and outcomes, facilitating knowledge sharing and team-wide learning.

* Developed pipelines for NLP and computer vision models (AWS Sagemaker), enabling the detection of * defect types and bringing transparency to the business, resulting in cost savings on non-relevant sources. * Created Power Bi Dashboards for analyzing road inspections and audits, providing valuable insights to * authorities and aiding in effective decision making. * Collaborated with cross-functional teams to gather requirements, design solutions, and implement scalable and efficient data-driven solutions for various projects.