AI/ML Engineer

New York City, New York

Machine Learning

Permanent

Our client, a fast-growing AI startup, is hiring an AI/ML Engineer to join their team in New York. The successful candidate will focus on building the core Copilot product, with ownership across the full ML lifecycle – from data pipelines and model training to embeddings, retrieval, serving and continuous iteration in production.

Responsibilities

Design, develop and deploy production-grade Machine Learning systems in Python, moving well beyond experimentation and notebooks.
Take ownership of key components of the NLP and LLM stack, including embeddings, retrieval pipelines, RAG architectures and targeted fine-tuning.
Build and iterate on recommendation and ranking models that surface who users should engage with and when.
Work hands-on with vector databases and similarity search to drive relationship intelligence.
Implement and maintain ML tooling for training, versioning, monitoring and evaluation.
Collaborate closely with founders, product and engineering to translate ideas into shipped, measurable product capabilities.
Integrate ML models into production APIs within a TypeScript / Nest.js–heavy environment.
Continuously optimise latency, cost and performance, including model routing, caching, distillation and quantisation.

Skillset

Minimum of 3 years of experience building and deploying production ML systems using Python, PyTorch or TensorFlow and scikit-learn.
Hands-on NLP and LLM experience, including HuggingFace Transformers, embeddings, sentence-transformers and RAG architectures.
Experience working with both proprietary models (e.g. OpenAI) and open-source LLMs (e.g. Llama, Mistral).
Strong foundation in classical machine learning, including classification, ranking, supervised and unsupervised learning, and XGBoost or LightGBM.
Experience with MLOps and infrastructure, such as experiment tracking, model versioning, Docker/Kubernetes, SageMaker or similar systems.
Practical experience with vector databases, including Pinecone, Qdrant, Weaviate or comparable platforms.
Strong software engineering fundamentals, with experience integrating ML models into reliable, production-grade systems.