For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
S
Sabastian S.

Sabastian S.

Principal AI Research Scientist and LLM Evaluation Lead – Open Mind Research Institute (Remote)

USA flagN/A, Usa

Key Skills

Software

Other
AppenAppen
AWS SageMakerAWS SageMaker
CrowdSourceCrowdSource
CloudFactoryCloudFactory
HiveMindHiveMind
HumanaticHumanatic
iMeritiMerit
MercorMercor

Top Subject Matter

LLM evaluation
computational correctness benchmarking
AI research

Top Data Types

DocumentDocument
TextText
VideoVideo

Top Task Types

Red TeamingRed Teaming
Bounding BoxBounding Box
SegmentationSegmentation
CuboidCuboid
Object DetectionObject Detection

Freelancer Overview

Principal AI Research Scientist and LLM Evaluation Lead – Open Mind Research Institute (Remote). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other, Internal, and Proprietary Tooling. Education includes Doctor of Philosophy, Massachusetts University of Technology (MIT) (2019) and Master of Science, Stanford University (2015). AI-training focus includes data types such as Document and labeling workflows including Evaluation, Rating, and Red Teaming.

Labeling Experience

Principal AI Research Scientist and LLM Evaluation Lead – Open Mind Research Institute (Remote)

OtherDocumentDocument

Led and managed LLM evaluation work for the OpenMind Computational Correctness Benchmark. Focused on designing evaluation frameworks and coordinating asynchronous research operations across multiple time zones. Oversaw delivery of evaluation methods intended to assess model computational correctness and related behaviors. • Directed a team of 8 senior researchers on evaluation framework design. • Managed remote research workflow across 5 time zones. • Conducted prompt/response assessment activities as part of LLM evaluation. • Supported benchmark engineering and correctness-focused evaluation deliverables.

2025 - Present

Software Engineer – QOC Innovations

DocumentDocumentRed TeamingRed Teaming

Built automated testing pipelines to improve reliability and vulnerability detection for an AI-adjacent platform. Led internal red-teaming initiatives to strengthen platform security posture. Produced documentation supporting mitigation and system hardening for risk reduction. • Developed/maintained automated testing pipelines to surface issues. • Conducted or coordinated red-teaming activities to identify security weaknesses. • Authored technical documentation for mitigations and hardening steps. • Supported continuous reliability and vulnerability detection efforts.

2024 - 2024

Education

M

Massachusetts University of Technology (MIT)

Doctor of Philosophy, Computer Science

Doctor of Philosophy
2016 - 2019
S

Stanford University

Master of Science, Computer Science

Master of Science
2012 - 2015

Work History

O

Open Mind Research Institute

Principal AI Research Scientist and LLM Evaluation Lead

N/A
2025 - Present
Q

Qoc Innovations

Software Engineer

N/A
2024 - Present