Lead NLP ML System Development
Budget: $3,000 – $5,000 USD
I’m ready to push a new natural-language-processing product from concept to production and need a senior Python engineer who can own the entire ML lifecycle. You’ll turn raw text data into a robust, scalable service that delivers measurable business value.
Here’s what I expect you to drive:
• Model research and rapid prototyping in TensorFlow, PyTorch, and Scikit-Learn, selecting the best architecture through solid experimentation and metrics.
• Data-pipeline design—cleaning, feature generation, versioning, and automated retraining—so every experiment is fully reproducible.
• Production-grade deployment: containerised inference endpoints, CI/CD, monitoring, and autoscaling on the cloud platform of your choice.
• Clear documentation and unit / integration tests that make ongoing maintenance straightforward for the wider engineering team.
Acceptance criteria
• An API or microservice that returns NLP predictions with latency under 100 ms and horizontal scalability demonstrated under load testing.
• End-to-end reproducibility: one command spins up infrastructure, trains the model, and deploys it.
• Coverage reports and README that allow another engineer to reproduce results in under an hour.
If you’re comfortable leading cross-functional discussions, enjoy balancing research depth with shipping deadlines, and have a track record of launching NLP systems at scale, let’s talk.
Here’s what I expect you to drive:
• Model research and rapid prototyping in TensorFlow, PyTorch, and Scikit-Learn, selecting the best architecture through solid experimentation and metrics.
• Data-pipeline design—cleaning, feature generation, versioning, and automated retraining—so every experiment is fully reproducible.
• Production-grade deployment: containerised inference endpoints, CI/CD, monitoring, and autoscaling on the cloud platform of your choice.
• Clear documentation and unit / integration tests that make ongoing maintenance straightforward for the wider engineering team.
Acceptance criteria
• An API or microservice that returns NLP predictions with latency under 100 ms and horizontal scalability demonstrated under load testing.
• End-to-end reproducibility: one command spins up infrastructure, trains the model, and deploys it.
• Coverage reports and README that allow another engineer to reproduce results in under an hour.
If you’re comfortable leading cross-functional discussions, enjoy balancing research depth with shipping deadlines, and have a track record of launching NLP systems at scale, let’s talk.