- 5 7 years experience in AI/ML Engineering Data Engineering or MLOps
- Strong Python expertise
- Experience with PyTorch/JAX
- Experience with distributed systems (K8s Ray Slurm)
- Hands-on MLflow or similar tracking systems
- Experience building production AI systems (not just notebooks)
- Strong debugging reliability and systems thinking mindset
- Clear documentation and collaboration skills
- Implement modular training pipeline components in PyTorch Lightning / JAX
- Build data ingestion preprocessing modules with PIT-correct dataset generation
- Enable seamless lake-based feature reads
- Integrate experiment tracking (MLflow / Weights & Biases)
- Capture registry-ready metadata & manage artifact export
- Contribute to tuning toolkit (sweeps baselines evaluation metrics)
- Generate structured experiment reports
- Write clear runbooks and experiment playbooks for reproducibility
5 7 years experience in AI/ML Engineering Data Engineering or MLOps Strong Python expertise Experience with PyTorch/JAX Experience with distributed systems (K8s Ray Slurm) Hands-on MLflow or similar tracking systems Experience building production AI systems (not just notebooks) Strong debugging r...
- 5 7 years experience in AI/ML Engineering Data Engineering or MLOps
- Strong Python expertise
- Experience with PyTorch/JAX
- Experience with distributed systems (K8s Ray Slurm)
- Hands-on MLflow or similar tracking systems
- Experience building production AI systems (not just notebooks)
- Strong debugging reliability and systems thinking mindset
- Clear documentation and collaboration skills
- Implement modular training pipeline components in PyTorch Lightning / JAX
- Build data ingestion preprocessing modules with PIT-correct dataset generation
- Enable seamless lake-based feature reads
- Integrate experiment tracking (MLflow / Weights & Biases)
- Capture registry-ready metadata & manage artifact export
- Contribute to tuning toolkit (sweeps baselines evaluation metrics)
- Generate structured experiment reports
- Write clear runbooks and experiment playbooks for reproducibility
View more
View less