Latest from Collinear AI

Deep dive into the product and research that power the enterprise
AI improvement flywheel

Deep dive into the product and research that power the enterprise AI improvement flywheel

RL Infrastructure for AI Agents: Why Environment-as-a-Service is the Missing Piece

Nazneen Rajani
Nov 19
2

Announcing Spider: a lightweight tool to craft post-training data recipes

Soumyadeep Bakshi
Nov 6
6

Collinear Newsletter #8 – Notes on Improving AI

Soumyadeep Bakshi
Nov 5
8

The case for simulations

Unlocking model uplift through better evaluations
Soumyadeep Bakshi and Grant Griffith
Oct 23
5

Through the Valley of Reasoning: What Small Models Teach Us About Learning

TraitBasis, inspired from mech intrep, gives high-fidelity user personas for comprehensive agent testing
Muyu, Tsach Mackey, Anand Kumar, and 2 others
Oct 9

Introducing Collinear Simulations: Steerable Personas for AI Agent Testing

TraitBasis, inspired from mech intrep, gives high-fidelity user personas for comprehensive agent testing
Meghana A Rajeev, Tsach Mackey, Muyu, and 2 others
Oct 7
2

Collinear Newsletter #7 | Notes on Improving AI

A deep dive into how leading companies are deploying AI agents reliably in production
Soumyadeep Bakshi and Nazneen Rajani
Oct 3
3

Introducing Curator Evals: A Benchmark for High-quality Post-training Data Curation

Muhammad Ali Shafique, Tsach Mackey, Meghana A Rajeev, Soumyadeep Bakshi, Muyu, Anand Kumar, and Nazneen Rajani
Sep 2
2

Collinear AI Now Available on Google Cloud Marketplace

Making safe, high-performing AI accessible through trusted enterprise infrastructure
Soumyadeep Bakshi
Aug 25
7

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
Aug 13
2

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, and 3 others
Aug 12
4

OpenAI's gpt-oss on LiveCodeBench: A Competitive Programming Deep Dive

OpenAI finally released an open-weights model, the gpt-oss. Their last open-weights release was GPT2, back in 2019!
Muyu, Tsach Mackey, Anand Kumar, and 2 others
Aug 6
2

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
Aug 5

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
Meghana A Rajeev, Jahnavi Jambholkar, and Nazneen Rajani
Jul 22
1

Data Curation: The secret sauce for enterprise AI excellence

How Collinear AI's Reward Models Transform Training Efficiency and Model Performance
Meghana A Rajeev, Tsach Mackey, and Nazneen Rajani
Jul 14
1

Newsletter #5 | 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
Jun 21
1

Judges as Data curators cut Post-training Time to Half

ServiceNow x Collinear
Nazneen Rajani, Meghana A Rajeev, and Tsach Mackey
Jun 10
1

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
May 16
2

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
May 7

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 Wintour. When they decided to create AI personas that could authentically represent these instructors, they faced a challenge that would make or break the entire experience.
Jahnavi Jambholkar, Nazneen Rajani, and Soumyadeep Bakshi
Apr 22
2

Issue #3 | 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
Apr 15

Issue #2 | Notes on Improving AI Performance

A deep dive into how leading companies are deploying AI agents reliably in production
Nazneen Rajani, Jahnavi Jambholkar, and Soumyadeep Bakshi
Mar 24
2

The Limitations of AI Evaluations

Why Enterprises Need a New Approach
Jahnavi Jambholkar, Soumyadeep Bakshi, and Nazneen Rajani
Mar 18
1

The AI Safety Gap

Why Traditional Models Fail in Modern Enterprises
Soumyadeep Bakshi and Nazneen Rajani
Feb 26
4

Issue #1 | Notes on Improving AI Performance

A behind-the-scenes look at building enterprise-grade, reliable AI, at scale
Soumyadeep Bakshi
Feb 12
7

Taming AI Agents: Why Your Butler Needs a Babysitter

They're everywhere!!!
Meghana A Rajeev, Nazneen Rajani, and Soumyadeep Bakshi
Jan 27
7

Collinear-Guard: Where Customization Meets Precision for Fine-Grained Evaluation and Feedback

LLM-as-Judge models are becoming more popular as evaluators in different areas, especially when it comes to aligning models with specific rules or principles using approaches like Reward Learning from AI Feedback (RLAIF).
Prapti Trivedi, Aditya Gulati, and Nazneen Rajani
Dec 9
6

CollinearGuard Nano: A High-Performance, Holistically-Evaluated, Lightning-Fast Moderation Judge

Revolutionizing Safety: Ultra-Low Latency, High-Throughput Violation and False Refusal Detection Like Never Before!
Aditya Gulati, Prapti Trivedi, and Nazneen Rajani
Dec 3
21

Veritas Reliability Judge: A Cookbook to Benchmark AI Judges on Financial Data

When you're working with high-stakes data, like financial information, ensuring the accuracy of AI-generated responses is
Meghana A Rajeev
Nov 18
3

Why AI Safety is existentially important, not optional

From virtual assistants that streamline our tasks to intelligent systems that drive innovation, AI's potential seems boundless. Yet, with this immense power comes a profound responsibility.
Jahnavi Jambholkar, Nazneen Rajani, Meghana A Rajeev, and 2 others
Oct 25
1

Introducing VERITAS: A Unified Approach to Reliability Evaluation

Veritas is a suite of Reliability Judges for Batch and Real-time use cases
Rajkumar Ramamurthy, Meghana A Rajeev, Nazneen Rajani, and Oliver Molenschot
Oct 24
1

A Guide to Creating Seed Conversational Data with Collinear

High-quality seed data can overcome many of the post-training challenges
Jahnavi Jambholkar, Nazneen Rajani, and Tanveesh Chaudhery
Oct 19
4

Think Before You Score: Self-Rationalizing Evaluators are State-of-the-Art for Fine-grained Evaluation

Finetuning on Rationales improves Judge Rationale and Score
Prapti Trivedi, Aditya Gulati, Oliver Molenschot, and 2 others
Oct 10
10

Collinear Flex Judge is Better Aligned than Few-shot Prompted GPT-4o

Accelerate Time to Production with Bespoke Quality Judge
Nazneen Rajani, Jahnavi Jambholkar, and Tanveesh Chaudhery
Oct 5
7

Ship smarter models, not bigger ones.

Collinear generates high signal post-training datasets and RL environments that make every release stronger.

Generate Data

Stop launch-and-pray AI.

Simulations catch failures before your customers do.

Try our Platform