Data Science Team Training is designed for data scientists, ML engineers, and MLOps practitioners who need to deploy and manage AI/ML workloads on PaletteAI. All tiers include hands-on lab exercises with GPU access, code examples, and post-training Q&A sessions.
Phase 3: Operate & Enable
Data Science Team Training
Training for data scientists and ML engineers on self-service provisioning, GPU workload submission, MLOps tooling, and model serving on PaletteAI.
Overview
What you get
- Structured curriculum tailored to data science and ML engineering teams
- Hands-on lab exercises with GPU-enabled PaletteAI environments
- Code examples and reference notebooks
- Post-training Q&A sessions with the instructor
- Completion certificate (all tiers) with optional skills assessment and proficiency badge
Quick facts
Phase 3: Operate & Enable
Cohort: Up to 12 participants
Prerequisite: Basic K8s awareness
Curriculum includes:
| ESSENTIALS | STANDARD | PREMIUM | |
|---|---|---|---|
| PaletteAI for ML workloads overview | X | X | X |
| GPU cluster provisioning and resource quotas | X | X | X |
| Deploying Jupyter / notebook environments | X | X | X |
| Model serving basics (KServe, Seldon) | X | X | X |
| Intro to Spectro Cloud Packs for ML | X | X | X |
| MLOps pipeline design (Kubeflow, Argo) | — | X | X |
| Multi-GPU scheduling and NUMA-aware placement | — | X | X |
| Model registry integration and versioning | — | X | X |
| Experiment tracking on PaletteAI | — | X | X |
| Data pipeline orchestration (Spark, Ray on K8s) | — | X | X |
| Advanced distributed training patterns | — | — | X |
| Custom operator development for ML workflows | — | — | X |
| Cost optimization for GPU clusters | — | — | X |
| Federated learning and multi-cluster model training | — | — | X |
| Production model monitoring and drift detection | — | — | X |
| Capstone: End-to-end MLOps pipeline build | — | — | X |
Feature comparison
| ESSENTIALS | STANDARD | PREMIUM | |
|---|---|---|---|
| Duration | 2 Days | 3 Days | 5 Days |
| Cohort | Up To 12 | Up To 12 | Up To 10 |
| Format | Virtual live | Virtual or on-site | On-site preferred |
| Labs | 3 guided labs | 6 labs + mini-project | 10 labs + capstone |
| Lab environment | Shared sandbox | Dedicated cluster | Dedicated + 30-day access |
| Instructor | Senior consultant | Senior + office hours | Lead data scientist + mentorship |
| Certification | Completion cert | + Skills assessment | + ISC proficiency badge |
How it works
1Pre-training
ISC reviews your team's skill levels and current tooling. GPU-enabled lab environments are provisioned and pre-configured with sample datasets.
2Training delivery
Instructor-led sessions combining theory with hands-on labs. Each tier builds on the previous, with increasing depth and complexity.
3Post-training
Q&A sessions with the instructor. Premium tier includes 30-day lab access and ongoing mentorship.
Deliverables
- Instructor-led training sessions (virtual or on-site)
- Hands-on lab exercises with GPU-enabled environments
- Code examples and reference notebooks
- Completion certificates for all participants
- Skills assessment results (Standard and Premium)
- ISC proficiency badge (Premium)
This is for you if...
Data Science Team Training is designed for data scientists, ML engineers, MLOps practitioners, and applied research teams who need to deploy and run AI/ML workloads on PaletteAI.
Basic Kubernetes awareness is recommended. No prior PaletteAI experience is required for Essentials.
Empower your data science team
Get your ML engineers productive on PaletteAI fast.