Machine Learning Engineer

TransRe•New York, NY

3d•Hybrid

About The Position

This role will be part of our TransRe Artificial Intelligent Team (TRAIT) and will be responsible for providing key Machine Learning deliverables to our global user base. TransRe has delivered first-class solutions to insurers worldwide since 1977, combining global reach with local decision-making. The company has built customer and broker relationships on years of trust, experience, and execution. Through its people, products, and partnerships, TransRe delivers the capacity and expertise necessary to contribute to the sustainable growth of prosperous communities worldwide. The company values Integrity, Respect, Performance, Entrepreneurship, and Customer Focus.

Requirements

Proven track record of building, scaling and productizing multiple machine learning models
Strong experience in the full life-cycle of machine learning models from initial theorizing to final implementation & support
High proficiency with working with LLMs- fine-tuning, RLHF, distillation, optimization
Experience working with sparse, high dimensional, tabular, and time series data
Python ml stack
PyTorch, TensorFlow, CUDA
Boosting and bagging algorithms
ML Optimization Techniques
Ensemble Stacking and Meta Learners

Responsibilities

Constructing machine learning models including data collection, normalization, and standardization, data pipeline construction, model selection and hyperparameter tuning, working ml systems that can add new data into ml model
Working on and researching the most recent LLM and GenAI models, safety, interpretability, and applications
Creating apps to present ml models and host them in the cloud or locally
Creating pipelines to query and retrieve and update data for existing applications to keep them updated
Supervising the scaling and management of the machine learning modeling ecosystem
Finding orthogonal data sets to supplement models and increase alpha
Staying abreast of new technology and machine learning methodologies and implementing them into the model building architecture
Working alongside (re)insurance domain experts to improve predictive aspects of their lines of business