ML Engineer (NLP / LLMs)

This profile represents the applied AI shape in highest demand right now: an ML engineer who came up through classical NLP and now ships LLM features that survive contact with real users.

Background

Six years in machine learning, starting with classical NLP and PyTorch models for text classification and extraction, then moving with the field into LLM-based systems. That trajectory matters: it means choosing between a fine-tuned small model, retrieval over a hosted model, or a plain heuristic based on the problem, not the hype. Recent work has been LLM features inside SaaS and healthtech products, deployed on AWS, where the healthtech side enforced real care about hallucinations and patient data.

Strengths

• Production LLM systems. Retrieval pipelines, prompt and model evaluation, and guardrails treated as core engineering.
• Model judgment. Knows when an API call, a fine-tune, or no model at all is the right answer.
• Evaluation habits. Builds measurement before shipping, so quality changes are observed rather than guessed.
• Sensitive-data caution. Healthtech experience with de-identification and constrained model outputs.

Typical engagement

Usually engaged to take an LLM feature from prototype to production, or to audit and harden one that is already live. MSK hours with good overlap for European teams. Currently at limited availability, so scoped projects and part-time arrangements are the realistic starting point.

This is a representative sample of the vetted pool. Request it and we confirm real, current availability for a developer with this background.