AI & ML interests

Private evals, RL environments and training data for AI agents on real creative jobs. First up, video editing.

Recent Activity

Organization Card
TensorTest. AI training data, evals, and RL environments.

AI training data, evals, and RL environments

Training data, evals, and RL environments for AI labs, robotics companies, and enterprises. Custom datasets, production traces, and domain-expert staffing.

VIDEO EDITING · TIMELINEBENCH

Agents say they’re done. Most aren’t.

93%Ended with the agent claiming success
14.0%Resolved [11.7, 16.4]

TimelineBench gave 16 agent setups 56 real editing jobs. Each is a whole job: raw footage, audio and a brief in, a finished cut out. A run resolves only if it passes hard checks on delivery, content and the brief, and a quality test calibrated on blind judgments by 43 professional editors.

Each square is one agent run. 836 of 896 runs ended with the agent claiming success; 125 (14.0%) resolved.
Each square is one agent run. Source: arXiv:2609.35143.

What we build

Private evals

Available now

We run your unreleased checkpoint or agent on our jobs and send a failure report. Results stay private.

RL environments

Scoped with your team

Our jobs, packaged for your sandbox, with our checks exposed as the reward.

Training and eval data

Commissioned

Pre-training and post-training data, custom datasets, and production traces scoped to your workflows.

Domain-expert staffing

We staff domain experts onto your team for data and evaluation work.

Tell us what you need. Terms are scoped per engagement.

The three founders wrote TimelineBench.

Two of us were Research Fellows at Microsoft Research India. Papers at NeurIPS, ACL Findings, EACL, Interspeech and CSCW.

  • Pavan Kalyan Tankala CEO
  • Nirmit Arora CTO
  • Gunin Gupta CBO
  • Y Combinator W26
TensorTest · San Francisco

models 0

None public yet