Job
Description
About Team Kensho Link is a machine learning service that allows users to map entities in their datasets with unique ID numbers drawn from S&P Globals world-class company database with precision and speed. Link started as an internal Kensho project to help S&P Global Market Intelligence and CapIQ integrate datasets more quickly into their platform. It uses ML based algorithm trained to return high quality links, even when the data inputs are incomplete or contain errors. Link leverages a variety of NLP and ML techniques to process and link millions of company entities in hours. As a team, we have expertise in classical ML algorithms and modern LLM-based tech stacks like RAG systems. About The Role As a Machine Learning Engineer, you will have end-to-end ownership of the Link application, driving its development and success. This role is perfect for you if you thrive on creating ML models and have a keen interest in the software engineering aspects of machine learning ( MLOps ). For example, y our responsibilities will include: Model deployment and optimization Debugging performance discrepancies between online and offline environments Ensuring feature synchronization Optimizing memory and compute resources Scaling ML systems for maximum efficiency Why Join Us? We are continuously expanding our portfolio of projects and are eager to find talented engineers who are excited about building and deploying state-of-the-art ML systems. We are seeking a mid-level Machine Learning Engineer who can help accelerate and refine our ML development cycle, setting new standards for prototyping, building, and maintaining cutting-edge ML solutions. What Youll Do Develop Advanced ML Models: Create innovative machine learning models to address complex business challenges and drive value. Enhance Model Performance: Identify and resolve performance gaps in existing models with creative and effective solutions. Leverage Unique Data: Work with proprietary unstructured and structured datasets, applying advanced NLP techniques to extract insights and build impactful solutions. Optimize Application Scaling: Efficiently scale ML applications to maximize compute resource utilization and meet high customer demand. Address Technical Debt: Proactively identify and propose solutions to reduce technical debt within the tech stack. Drive the ML Lifecycle: Engage in all phases of the ML lifecycle, from problem framing and data exploration to model deployment and production monitoring, ensuring continuous improvement. Collaborate Across Teams: Partner with cross-functional teams, including Data, Product Management, Design, and Engineering, to ensure smooth operations and contribute to the future product vision. Lead ML Projects: Oversee the development of core capabilities by scoping and planning ML projects effectively. Enhance User Experiences: Collaborate with Product and Design teams to develop ML-based solutions that enhance user experiences and align with business goals. Who You Are Bachelor's degree or higher in Computer Science, Engineering, or a related field. 3+ years of significant, hands-on industry experience with machine learning, natural language processing (NLP), and information retrieval systems, including designing, shipping, and maintaining production systems. Strong proficiency in Python. Proven experience building ML pipelines for data processing, training, inference, maintenance, evaluation, versioning, and experimentation. Demonstrated effective coding, documentation, collaboration, and communication habits. Strong problem-solving skills and a proactive approach to addressing challenges. Ability to adapt to a fast-paced and dynamic work environment. [optional] Experience developing search or recommender systems [optional] Experience working with databases and other datastores [optional] Experience working with machine learning libraries/frameworks for Large Language Model (LLM) orchestration, such as Langchain, Semantic Kernel, LLamaIndex, etc. Technologies We Love Traditional ML: SKLearn, XGBoost, LightGBM ML/Deep Learning: PyTorch, Transformers, HuggingFace, LangChain Deployment: Airflow, Docker, Kubernetes, Jenkins, AWS EDA/Visualization : Pandas, Matplotlib, Jupyter, Weights & Biases Tools/Toolkits: DVC, MosaicML, NVIDIA NeMo, LabelBox Techniques : RAG, Prompt Engineering, Information Retrieval, Data Embedding Datastores : Postgres, OpenSearch, SQLite, S3
