AI Developer Certificate

Directory / Google / TensorFlow

TensorFlow Developer Certificate

A hands-on coding exam in classical deep learning with TensorFlow. It has no generative AI content.

What it signals

Practical assessment: Performance-based labs or a practical build (4 of 4) Practical assessment
Performance-based labs or a practical build. Practical model-building exam (historical, not re-verified).
Identity assurance: Proctored, arrangements vary by provider (3 of 4) Identity assurance
Proctored, arrangements vary by provider. No longer offered.
Depth: Associate (2 of 4) Depth
Associate. Associate level.
– Experience expected
Not rated. No official recommendation verified.
Portability: Mixed (2 of 4) Portability
Mixed. Deep learning concepts transfer but the exam was framework-specific.
Verifiability: Public badge or registry (4 of 4) Verifiability
Public badge or registry. Public Certificate Network directory still linked from the official page.

For hiring managers

  • Cannot be earned any more. Any certificate is from May 2024 or earlier; all will have expired by about mid-2027, and those earned before Sept 2023 are already expired.
  • No replacement has been launched despite the 'evaluating next step' wording, which has been on the page for over two years.
  • It was a hands-on coding exam in classical deep learning (no LLM/gen AI content), so it remains decent evidence of past practical skill.

Verifying it

Official page still links to a TensorFlow Certificate Network to 'find developers who have passed the exam'.

What it covers

By topic

GenAI apps: Not covered (0 of 4) GenAI apps
Not covered. Pre-LLM-era content.
Agents: Not covered (0 of 4) Agents
Not covered. Not covered.
Classical ML: Core focus (over 35%) (4 of 4) Classical ML
Core focus (over 35%). Entirely about building and training neural network models (historical blueprint, not re-verified).
Operations: Not covered (0 of 4) Operations
Not covered. Not covered as far as known.
Data: Touched on (under 10%) (1 of 4) Data
Touched on (under 10%). Some data loading/preprocessing for training (not re-verified).
Responsible AI: Not covered (0 of 4) Responsible AI
Not covered. Not covered.

Exam details

Who the provider says it is for. Developers/data scientists demonstrating TensorFlow model-building skill.

Format
historically a practical exam: build and train TensorFlow models in an IDE (not re-verified)
Delivery
No longer offered.
Code or hands-on work
Historically yes - candidates wrote TensorFlow code to train models (not re-verified from an official source in this session).
Renewal
None - exam closed, credentials cannot be renewed
Price notes
Historically USD 100 (from memory/third-party, not re-verified)

Not confirmed

These details were not available from an official source at the time of research.

  • exact closure date (1 May vs 31 May 2024)
  • historical price, duration, format, blueprint (not on the current official page)
  • whether Certificate Network directory is still functional
Sources (1)