Skip to main content![The Mosaic Research Team]()
![OfficeQA: A grounded reasoning benchmark for testing enterprise AI]()
![ai_parse_document]()
![]()
![Judging with Confidence: Meet PGRM, the Promptable Reward Model]()
![Agent Learning from Human Feedback (ALHF): A Databricks Knowledge Assistant Case Study]()
![The power of RLVR: Training a Leading SQL Reasoning Model on Databricks]()
![Test-time Adaptive Optimization (TAO)]()
![Introducing MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs]()
The Mosaic Research Team
The Mosaic Research Team's posts
Mosaic Research
December 9, 2025/12 min read
Introducing OfficeQA: A Benchmark for End-to-End Grounded Reasoning
Product
November 11, 2025/5 min read
PDFs to Production: Announcing state-of-the-art document intelligence on Databricks
Mosaic Research
September 24, 2025/12 min read
Building State-of-the-Art Enterprise Agents 90x Cheaper with Automated Prompt Optimization
Mosaic Research
August 12, 2025/10 min read
Judging with Confidence: Meet PGRM, the Promptable Reward Model
Mosaic Research
August 4, 2025/7 min read
Agent Learning from Human Feedback (ALHF): A Databricks Knowledge Assistant Case Study
Mosaic Research
July 30, 2025/4 min read
The Power of RLVR: Training a Leading SQL Reasoning Model on Databricks
Mosaic Research
March 25, 2025/8 min read
TAO: Using test-time compute to train efficient LLMs without labeled data
Mosaic Research
May 5, 2023/18 min read