Collinear AI’s Blog
Subscribe
Sign in
Home
Archive
About
Latest
Top
Discussions
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
Aug 6
•
Muyu
,
Tsach Mackey
,
Anand Kumar
,
Meghana A Rajeev
, and
Nazneen Rajani
1
Share this post
Collinear AI’s Blog
OpenAI's gpt-oss on LiveCodeBench: A Competitive Programming Deep Dive
Copy link
Facebook
Email
Notes
More
3
Newsletter #6 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
Aug 5
•
Nazneen Rajani
and
Soumyadeep Bakshi
Share this post
Collinear AI’s Blog
Newsletter #6 | Notes on Improving AI Performance
Copy link
Facebook
Email
Notes
More
July 2025
Cats confuse LRMs: Exposing blind spots in SOTA Models
Irrelevant, universal phrases like "Interesting fact: Cats sleep most of their lives" appended to math problems can break AI models
Jul 22
•
Meghana A Rajeev
,
Jahnavi Jambholkar
, and
Nazneen Rajani
2
Share this post
Collinear AI’s Blog
Cats confuse LRMs: Exposing blind spots in SOTA Models
Copy link
Facebook
Email
Notes
More
Data Curation: The secret sauce for enterprise AI excellence
How Collinear AI's Reward Models Transform Training Efficiency and Model Performance
Jul 14
•
Meghana A Rajeev
,
Tsach Mackey
, and
Nazneen Rajani
Share this post
Collinear AI’s Blog
Data Curation: The secret sauce for enterprise AI excellence
Copy link
Facebook
Email
Notes
More
June 2025
Newsletter #5 | Notes on Improving AI Performance
Hi AI innovators,
Jun 20
•
Jahnavi Jambholkar
,
Nazneen Rajani
, and
Soumyadeep Bakshi
1
Share this post
Collinear AI’s Blog
Newsletter #5 | Notes on Improving AI Performance
Copy link
Facebook
Email
Notes
More
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
May 15
•
Jahnavi Jambholkar
,
Nazneen Rajani
, and
Soumyadeep Bakshi
2
Share this post
Collinear AI’s Blog
Gaming the System: Goodhart’s Law Exemplified in AI Leaderboard Controversy
Copy link
Facebook
Email
Notes
More
Judges as Data curators cut Post-training Time to Half
ServiceNow x Collinear
May 14
•
Nazneen Rajani
,
Meghana A Rajeev
, and
Tsach Mackey
3
Share this post
Collinear AI’s Blog
Judges as Data curators cut Post-training Time to Half
Copy link
Facebook
Email
Notes
More
Newsletter #4 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
May 7
•
Jahnavi Jambholkar
,
Nazneen Rajani
, and
Soumyadeep Bakshi
Share this post
Collinear AI’s Blog
Newsletter #4 | Notes on Improving AI Performance
Copy link
Facebook
Email
Notes
More
April 2025
From worlds best pros to AI personas: The MasterClass journey
MasterClass has redefined personalized learning by providing access to the world's best minds – from culinary master Gordon Ramsay to fashion icon Anna…
Apr 22
•
Jahnavi Jambholkar
,
Nazneen Rajani
, and
Soumyadeep Bakshi
2
Share this post
Collinear AI’s Blog
From worlds best pros to AI personas: The MasterClass journey
Copy link
Facebook
Email
Notes
More
Issue #3 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
Apr 15
•
Jahnavi Jambholkar
,
Nazneen Rajani
, and
Soumyadeep Bakshi
Share this post
Collinear AI’s Blog
Issue #3 | Notes on Improving AI Performance
Copy link
Facebook
Email
Notes
More
March 2025
Issue #2 | Notes on Improving AI Performance
A deep dive into how leading companies are deploying AI agents reliably in production
Mar 24
•
Nazneen Rajani
,
Jahnavi Jambholkar
, and
Soumyadeep Bakshi
2
Share this post
Collinear AI’s Blog
Issue #2 | Notes on Improving AI Performance
Copy link
Facebook
Email
Notes
More
The Limitations of AI Evaluations
Why Enterprises Need a New Approach
Mar 18
•
Jahnavi Jambholkar
,
Soumyadeep Bakshi
, and
Nazneen Rajani
1
Share this post
Collinear AI’s Blog
The Limitations of AI Evaluations
Copy link
Facebook
Email
Notes
More
Share
Copy link
Facebook
Email
Notes
More
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts
