• Lead the transition of machine learning models from theoretical prototypes into scalable, high-performance production systems. • Architect and deploy ML solutions utilizing AWS ECS (Elastic Container Service) for containerized workloads and AWS Lambda for serverless, event-driven inference pipelines. • Optimize PyTorch models for production deployment by converting them to ONNX formats. • Apply advanced inference optimization techniques (quantization, pruning, ONNX Runtime) and memory-efficient attention mechanisms like Flash Attention to minimize latency and maximize throughput. • Champion infrastructure best practices for machine learning systems, establishing reliable CI/CD pipelines, and ensuring robust, secure, and reproducible deployments across the AWS ecosystem. • Design, develop, and evaluate algorithms that generate descriptive, diagnostic, predictive, and prescriptive insights from both structured and unstructured data. • Write clean, efficient, and well-tested code. Complete rigorous testing, debugging, and documentation to ensure seamless installation and long-term maintenance. • Actively participate in research discussions, requirements gathering, and system design alongside domain experts to build tailored scoring and ML solutions.

Role Description As a Senior Machine Learning Engineer at Orita, you will: - Build and Productionize Models: Design, train, and deploy models that directly power our marketing-focused products, primarily for marketing use cases. - Develop Scalable ML Infrastructure: Architect and maintain robust, scalable, MLOps pipelines to ensure reliable training, serving, and monitoring of models in production. - Experiment & Optimize: Drive continuous improvement using A/B testing, uplift modeling, causal inference, and other advanced experimentation frameworks to validate and refine model performance. - Collaborate & Mentor: Work closely with cross-functional teams, including the CEO and CTO, to align on product goals and foster best practices for machine learning and data engineering across the organization. Qualifications - 5+ years of full-time software engineering experience, including at least 3 years working on ML systems. - Deep knowledge of modern machine learning algorithms (tree-based methods, deep learning architectures, transformers/LLMs). - Hands-on experience with PyTorch, TensorFlow, XGBoost or equivalent frameworks. - Feature engineering using aggregations, embeddings, and sub-models. - Track record building production-scale ML infrastructures, ideally using GCP (Vertex AI, KubeFlow, BigQuery, etc.). - Familiarity with CI/CD, containerization (Docker/Kubernetes), and distributed training (Spark, Ray, Dask, etc.). - Experience iterating models in a production environment is a must. - Strong proficiency in Python (numpy, pandas, etc.). - Experience with scalable data processing (Spark, Ray, BigQuery). - Job orchestration (Airflow). - Comfortable with advanced experimentation techniques. - Understanding of performance measurement in real-world deployments. - Comfortable wearing many hats—data wrangling, model development, deployment, monitoring, and performance optimization. - Excellent communication—able to explain complex ML concepts to non-technical stakeholders. - Self-starter mentality with the ability to own projects from ideation to deployment, picking up and learning new technologies as needed. Requirements - Familiarity with marketing technology or ads is a strong plus. - Experience with experimental design and methods such as causal inference or uplift modeling. - Exposure to modeling with LLMs and modern AI tooling. - Productionizing Reinforcement Learning and Bandit algorithms. - Ph.D in a technical field. - Experience in a fast-paced or startup environment. - You live in or near New York City. Most of us work in EST. Benefits - Impact: Join a lean, agile team shaping the future of ML for leading global brands. - Growth: Work directly with industry veterans with strong academic and professional backgrounds. - Innovation: Experiment with the latest ML models, from tree-based methods to cutting-edge LLMs. - Culture: We value ownership, iteration, and continuous learning—everyone’s voice matters.

• Author the AI/ML Architecture Design Document • Recommend an accuracy threshold for the catalog mapping engine. • Build the 4-tier catalog code mapping engine for SAP catalog types. • Implement the AI anomaly detection module. • Design and implement the "Train Model" feedback loop where operator corrections retrain the model. • Wrap the model as a microservice for the main app.

• Develop and implement advanced search solutions • Lead the development and optimization of search algorithms and systems utilizing technologies such as Elastic/OpenSearch • Fine-tune search relevance using machine learning techniques • Explore and implement cutting-edge technologies including Agentic Solutions, RAG, and LLMs • Analyze search data and user behavior to inform optimization strategies • Collaborate with cross-functional teams to understand search requirements and deliver results • Develop and evaluate analytical models to investigate search-related problems • Utilize strong programming skills in Python to examine and interpret search data