Collinear AI’s Blog

Collinear AI’s Blog

Home
Archive
About
You Can’t Hire Your Way to Model Alignment
Why the Global AI talent shortage Is undermining enterprise model alignment, and what you can do instead
 • 
Marc Moring
, 
Soumyadeep Bakshi
, and 
Nazneen Rajani
Leveling the Playing Field: Livecodebench’s Big Bug Fix
Three major fixes that reshaped competitive coding scores and why your numbers may look very different now
 • 
Muyu
, 
Tsach Mackey
, 
Muhammad Ali Shafique
, 
Anand Kumar
, 
Meghana A Rajeev
, and 
Nazneen Rajani
OpenAI's gpt-oss on LiveCodeBench: A Competitive Programming Deep Dive
tl;dr: the gpt-oss-20b is a strong model for competitive coding but is >3x sample inefficient compared to deespseek-r1-0528
 • 
Muyu
, 
Tsach Mackey
, 
Anand Kumar
, 
Meghana A Rajeev
, and 
Nazneen Rajani
3
Newsletter #6 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
 • 
Nazneen Rajani
 and 
Soumyadeep Bakshi

July 2025

June 2025

May 2025

Gaming the System: Goodhart’s Law Exemplified in AI Leaderboard Controversy
How the race to the top in AI benchmarks is leading to specialized optimization at the expense of real-world performance
 • 
Jahnavi Jambholkar
, 
Nazneen Rajani
, and 
Soumyadeep Bakshi
Judges as Data curators cut Post-training Time to Half
ServiceNow x Collinear
 • 
Nazneen Rajani
, 
Meghana A Rajeev
, and 
Tsach Mackey
Newsletter #4 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
 • 
Jahnavi Jambholkar
, 
Nazneen Rajani
, and 
Soumyadeep Bakshi

April 2025

© 2025 CollinearAI
Privacy ∙ Terms ∙ Collection notice
Start writingGet the app
Substack is the home for great culture