Elite experts for frontier AI, from Latin America.
Entor deploys vetted, English-fluent researchers across Colombia to label data, run RLHF and evaluations, and complete the expert tasks that make your models better.
Purpose-built for the teams training frontier models.
Problem
Better models still need better humans.
Frontier models have scraped the easy internet. The gains now come from human work: labeled data, ranked preferences, careful evaluations, and expert judgment on the hard cases. That work is scarce, slow, and expensive, and the best English-fluent experts are hard to reach. Entor solves it with a vetted network of researchers across Colombia, matched to your domain and working in your timezone.
We put Latin America’s best minds to work on the world’s hardest AI problems.
- Elite, English-fluent experts are scarce and expensive to hire in the US.
- Every task is attributable to a named, vetted expert, not anonymous piecework.
- Colombian talent works in US-timezone overlap, at a fraction of the cost.
What we do
Expert human work, on demand.
We deploy vetted experts to do the human work your models need, matched by domain and delivered in your timezone, across law, STEM, medicine, finance, code, and linguistics.
Data labeling & annotation
Classification, ranking, transcription, and structured annotation across text, code, and multimodal data, produced by domain experts.
RLHF & preference data
Pairwise and ranked human preferences with expert rationales, turning subjective judgment into scalable, attributable reward signals.
Evals, benchmarks & safety
Expert-built evaluations, red-teaming, and safety checks that show where your model breaks down and by how much.
Agentic & task completion
Experts operating tools, browsers, and codebases to demonstrate and complete real multi-step tasks, step-annotated and graded.
The network
A vetted expert network, across Colombia.
Entor recruits undergraduate to PhD researchers from Colombia’s top universities: English-fluent, working in US-timezone overlap, and vetted before any live work.
- Universidad de los AndesBogotá
- Universidad Nacional de ColombiaBogotá
- Pontificia Universidad JaverianaBogotá
vetted expert contributors
partner universities in Bogotá
target inter-rater agreement κ
daily timezone overlap
Research
Our work starts with research.
We ask where exactly models break down in real professional work, and why. Every domain has its own failure patterns, and we build the data that fixes them.
Where models still trail human experts
Spanish-language and Latin-American-law reasoning are where frontier models fall behind most, and why closing that gap matters.
Blog · Apr 28, 2026
The human baseline, explained
How we think about measuring expert performance against frontier models on contamination-controlled evaluations.
Blog · Apr 9, 2026
Inside the house standard
Scope, calibrate, author, two-pass blind review. The process we design every dataset around.
Blog · Mar 31, 2026
From the founder
“I started Entor to connect frontier AI labs with the best minds in Latin America, English-fluent researchers, undergraduate to PhD, recruited from Colombia’s top universities. The reasoning that separates a good answer from a correct one lives inside experts. We commission it, grade it, and stand behind every datum with a name.”
Zinnie Zhang
CEO & Founder
J.D. Candidate, Harvard Law School · B.S. Computer Science, Vanderbilt University
Founders
The people building Entor
A small team in Bogotá, working with Latin America’s best researchers.
Zinnie Zhang
CEO & Founder
Started Entor to connect frontier AI labs with the best minds in Latin America. Harvard Law and Vanderbilt CS; now building the expert-data house she wished existed.
J.D. Candidate, Harvard Law School · B.S. Computer Science, Vanderbilt University
Kevin Gámez
Co-founder
Builds and operates Entor’s platform and engineering, from Bogotá.
Bogotá, Colombia
For experts
Put your expertise to work.
Join Entor’s network of vetted experts and get matched to projects from frontier AI labs: labeling, ranking, evaluations, red-teaming, and expert tasks in your field. Remote, US-timezone, paid per project, and credited by name. We recruit undergraduate to PhD talent from Los Andes, Nacional, and Javeriana.
Join as an expertFields we recruit
- STEM & Code
- Law & Finance (ES / LatAm)
- Medicine & Life Sciences
- Linguistics & Spanish
- Safety & Red-teaming
- Agentic / Computer-use
Book a call.
Tell us what your model needs. We will scope a pilot and match you with the right experts within days.