← Back to Skills

TensorFlow/Keras Engineer

Expert in TensorFlow and Keras — model building, training, TF Serving, TF Lite, TFX pipelines, and production deployment on Google Cloud.

Gold
v1.0.00 activationsAI & Machine LearningTechnologyexpert

SupaScore

84.5
Research Quality (15%)
8.4
Prompt Engineering (25%)
8.6
Practical Utility (15%)
8.5
Completeness (10%)
8.2
User Satisfaction (20%)
8.5
Decision Usefulness (15%)
8.3

Best for

  • Building TensorFlow Serving REST/gRPC endpoints for production model inference
  • Optimizing tf.data pipelines with caching, prefetching, and parallel processing
  • Implementing distributed training strategies across multi-GPU and multi-node setups
  • Converting models to TF Lite for mobile/edge deployment with quantization
  • Creating TFX pipelines for end-to-end ML workflows from data ingestion to model serving

What you'll get

  • Complete tf.data pipeline code with interleave, cache, and prefetch optimizations including performance benchmarks
  • Multi-GPU distributed training implementation using MirroredStrategy with proper data sharding and gradient aggregation
  • TensorFlow Serving deployment configuration with batching, model versioning, and monitoring setup
Not designed for ↓
  • ×PyTorch model development or conversion to ONNX format
  • ×MLOps platform architecture beyond TensorFlow ecosystem tools
  • ×Cloud infrastructure provisioning or Kubernetes cluster management
  • ×Data preprocessing workflows outside of tf.data and TFX Transform
Expects

Clear problem definition with model architecture requirements, data characteristics, and deployment constraints (latency, throughput, edge vs cloud).

Returns

Production-ready TensorFlow code with optimized data pipelines, model architectures, training loops, and deployment configurations including performance benchmarks.

Evidence Policy

Standard: no explicit evidence policy.

tensorflowkerastf-servingtf-litetfxdeep-learninggoogle-cloud

Research Foundation: 6 sources (4 official docs, 2 paper)

This skill was developed through independent research and synthesis. SupaSkills is not affiliated with or endorsed by any cited author or organisation.

Version History

v1.0.02/15/2026

Initial version

Works well with

Need more depth?

Specialist skills that go deeper in areas this skill touches.

Common Workflows

TensorFlow Production ML Pipeline

End-to-end workflow from model development through production serving with observability

Activate this skill in Claude Code

Sign up for free to access the full system prompt via REST API or MCP.

Start Free to Activate This Skill

© 2026 Kill The Dragon GmbH. This skill and its system prompt are protected by copyright. Unauthorised redistribution is prohibited. Terms of Service · Legal Notice