Articles by benocodes
25

A real-world benchmark for AI code review (qodo.ai)

2

The 'weird' things that happened when ClickHouse replaced C++ with Rust (thenewstack.io)

17

Introducing Qodo Gen CLI: Build and Run Coding Agents Anywhere in the SDLC (qodo.ai)

3

Google Donates the Agent2Agent Protocol to the Linux Foundation (thenewstack.io)

17

How to avoid P hacking (nature.com)

3

Skunk cabbages and other smelly plants brew their foul odour (nature.com)

1

Migrating Large-Scale Interactive Compute Workloads to K8s Without Disruption (uber.com)

4

Fixrleak: Fixing Java Resource Leaks with GenAI (uber.com)

3

Advancing Invoice Document Processing at Uber Using GenAI (uber.com)

1

How generative AI is transforming developer workflows at Amazon (amazon.com)

2

Foundation Model for Personalized Recommendation (netflixtechblog.com)

19

How Airbnb measures listing lifetime value (medium.com/airbnb-engineering)

2

Ethically sourced "spare" human bodies could revolutionize medicine (technologyreview.com)

3

DeepSeek-V3 runs at 20 tokens/s on Mac Studio, and that's a nightmare for OpenAI (venturebeat.com)

1

The Situation at Columbia (columbia.edu)

2

RNA function follows form – why is it so hard to predict? (nature.com)

2

A Prenatal Test of the Fetus Turns Up Cancers in Pregnant Mothers (scientificamerican.com)

8

The Greenland Ice Sheet is fracturing faster than expected (nature.com)

5

Solving key challenges in AI-assisted code reviews (qodo.ai)

1

Considerations for making a tree view component accessible (github.blog)

34

Cloud Efficiency at Netflix (netflixtechblog.com)

1

Introducing Configurable Metaflow (netflixtechblog.com)

2

How Will We Know We're Not Alone? (quantamagazine.org)

1

How close is AI to human-level intelligence? (nature.com)

219

Model Context Protocol (anthropic.com)

16

WebSockets cost us $1M on our AWS bill (recall.ai)

1

Exploring Gen AI: Copilot's new multi-file editing (martinfowler.com)

1

Dental evidence for extended growth in early Homo from Dmanisi (nature.com)

43

Netflix's Distributed Counter Abstraction (netflixtechblog.com)

1

Debate May Help AI Models Converge on Truth (quantamagazine.org)

1

Vision Language Models Are In-Context Value Learners (generative-value-learning.github.io)

2

There Are Three Types of Twilight (scientificamerican.com)

3

How the Brain Summons Deep Sleep to Speed Healing (scientificamerican.com)

1

Gene-editing tool will help the world cope with climate change (technologyreview.com)

33

Support for Claude Sonnet 3.5, OpenAI O1 and Gemini 1.5 Pro (qodo.ai)

119

Chain-of-thought can hurt performance on tasks where thinking makes humans worse (arxiv.org)

108

LLMs know more than they show: On the intrinsic representation of hallucinations (arxiv.org)

1

Custom LLM as a Judge to Detect Hallucinations with Braintrust (cookbook.openai.com)

1

Improving Conversational AI at Airbnb (medium.com/airbnb-engineering)

1

Agents Thinking Fast and Slow: A Talker-Reasoner Architecture (arxiv.org)

2

Sparse Crosscoders for Cross-Layer Features and Model Diffing (transformer-circuits.pub)

3

Getting Claude Computer Use agent to spin up another agent in its VM (twitter.com/gavriel_cohen)

2

OpenAI's approach to AI and national security (openai.com)

1

OpenAI scientist: '20 seconds of thinking worth 100,000x more data' (venturebeat.com)

1

Claude Computer Use agent spins up another agent in its VM (twitter.com/gavriel_cohen)

77

Drasi: Microsoft's open source data processing platform for event-driven systems (github.com/drasi-project)

1

How Uber Optimizes LLM Training (uber.com)

4

Solving complex problems with OpenAI o1 models (openai.com)

10

The open future of networking hardware for AI (fb.com)

1

Port raises $35M for its end-to-end Internal Developer Portal (getport.io)

1

OpenAI's Swarm AI agent framework: Routines and handoffs (venturebeat.com)

48

AlphaCodium outperforms direct prompting of OpenAI's o1 on coding problems (qodo.ai)

32

Avoiding a Geopolitical open-source Apocalypse (thenewstack.io)

1

Building a Global Caching System at Netflix (infoq.com)

144

Upgrading Uber's MySQL Fleet (uber.com)

1

QueryGPT – Natural Language to SQL Using Generative AI (uber.com)

1

Driving a Project: Intern Edition (slack.engineering)

1

Qodo raises $40M Series A for quality-first code generation and testing (techcrunch.com)

1

Using the Pinecone vector database in .NET (infoworld.com)

10

Protocol Buffer Design: Principles and Practices for Collaborative Development (lyft.com)