AI/ML Engineer

Paroksya; United States; Hybrid; USD 120,000 - 170,000; Full time

Job summary

Paroksya is seeking a hands-on ML engineer to build production-grade models for real client products in the United States. You will design and train models for client-specific use cases and own the production inference pipeline from data to deployment, ensuring reliability under real traffic.

You'll collaborate with data engineering and stakeholders to explain model behavior, monitor performance, and iterate with low-latency inference expertise. Prior experience with ML deployment is essential.

Qualifications

• 3+ years shipping ML models to production.
• Strong Python, comfortable with PyTorch or TensorFlow.
• Understanding of MLOps: deployment and monitoring.

Responsibilities

• Design and train models for client-specific use cases.
• Build the production inference pipeline, not just the notebook.
• Work with data engineering to ensure clean, reliable training data.
• Explain model behavior and limitations clearly to non-technical stakeholders.

Skills

Python

MLOps

Production ML

PyTorch

TensorFlow

Model deployment

Tools

AWS SageMaker

Real-time inference systems

Job description

Job Summary

Build production ML systems for real client products: fraud detection, recommendation engines, predictive models.

You’ll work on ML systems that ship to production and handle real traffic, not research prototypes. Recent work includes a real-time fraud detection engine and a predictive maintenance platform running across 6 factories.

Responsibilities

• Design and train models for client-specific use cases
• Build the production inference pipeline, not just the notebook
• Work with data engineering to ensure clean, reliable training data
• Explain model behavior and limitations clearly to non-technical stakeholders

Requirements

• 3+ years shipping ML models to production
• Strong Python, comfortable with PyTorch or TensorFlow
• Understanding of MLOps: not just model training but deployment and monitoring

Nice to Have

• Experience with real-time/low-latency inference systems
• AWS SageMaker or similar experience