TensorTest
Evals, RL environments and training data for creative tasks
TensorTest (formerly Ritivel) builds evaluations, reinforcement-learning environments, and expert training data for AI labs developing creative agents, starting with video editing. An agent can produce a valid video file without delivering an edit that meets the client’s brief. Labs need reliable ways to measure creative quality, identify failures, and give agents feedback they can learn from. Our first benchmark, TimelineBench, gives agents client video-editing jobs with raw footage, audio, and a brief. It evaluates the finished work using a verifier calibrated on thousands of blind judgments from professional editors. Across 16 agents we tested, the best passed only 26.8% of jobs. Most failures came down to creative craft rather than broken output. We offer labs private evaluations to identify where their models fail, training environments that use our verifier’s scores as rewards, and expert-created training data, including professional editing demonstrations, preference pairs, and grading rubrics. These offerings connect evaluation to training: identify what an agent gets wrong, provide feedback on its attempts, and supply examples of better work. Our team comes from IIT Bombay and has trained language models and built agentic systems at Microsoft Research.
- Website
- tensortest.com
- City
- San Francisco
- Country
- United States
- YC batch
- Winter 2026
- Team
- 3 people
- Stage
- Early stage
- AI-native
- Yes
More AI & models in San Francisco
AnthropicLab$965bnMafia
Safety-focused lab building the Claude models.
OpenAILab$852bn
Frontier model lab behind GPT and ChatGPT.
Cognition AI$48bn
Artificial Intelligence. Valued at $48bn (September 2026).
Scale AI$29bnHiring
Data-centric infrastructure to accelerate the development of AI
Perplexity$20bnMafia
Answer engine that cites its sources.
Faire$12.6bnHiring
The global online platform empowering independent retail.
Claim this profile
Free and instant: prove you run tensortest.com and edit the profile yourself.
- Your logo, office address and founder photos
- The ✓ Verified mark next to TensorTest across the site
- Open roles listed on the Hiring board
- A followed link back to your website
A rank card with TensorTest's standing on Labspie, or a simple badge. Both link to this profile and update themselves.
