Skip to content
View AndresCotton's full-sized avatar
4 followers · 1 following

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Pull Shark

Achievements

Achievement: Pair ExtraordinaireAchievement: YOLOAchievement: QuickdrawAchievement: Pull Shark

Block or report AndresCotton

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AndresCotton/README.md

Andrés Cotton

Technical AI Safety Research Manager at Meridian Cambridge

About Me

I work on AI safety and LLM systems, focusing on agentic evals and alignment. I've managed technical research teams and engineers, and like building things from scratch. My background spans machine learning, generative ai, neuroscience, and linguistics.

Current Focus:

  • Agentic AI systems and goal-oriented tasks
  • Multi-agent coordination and control & alignment evals

Recent Work:

  • LLM Engineer at Latent
  • 📄 The Greatest Good Benchmark (EMNLP 2024) - measuring LLM alignment with utilitarian principles
  • 🎓 MSc dissertation on AI agent failures in open-ended coordination tasks
  • 🔧 Building evaluation frameworks using Inspect AI, LangGraph, and PyTorch

Background

  • Education: MSc Speech & Language Processing (University of Edinburgh), BA Linguistics & Neuroscience (UBA)
  • Previously: Director of Prompt Engineering at Cience
  • Co-founder: Chevening in AI Network (CHAIN)

📫 Reach me: namelastname @gmail.com |

Popular repositories Loading

  1. agentic-eval-multi-participant-coordination agentic-eval-multi-participant-coordination Public

    AI agent evaluation framework for multi-participant coordination tasks. Built with LangGraph, custom MCP tools, and LLM-as-a-Judge evaluation. MSc dissertation project (University of Edinburgh, 2025).

    Python 1 1

  2. octotools-with-mcp octotools-with-mcp Public

    Forked from octotools/octotools

    OctoTools with MCP integration

    Python 1

  3. AndresCotton AndresCotton Public

    Config files for my GitHub profile.

  4. rational_numbers rational_numbers Public

    Datasets used for the paper **Mental representations of rational numbers: a Massive Online Experiment**.

  5. RioPlatenseSpanish_Neurosurgery_Dataset RioPlatenseSpanish_Neurosurgery_Dataset Public

    1

  6. Reading-Comprehension-on-Smartphones Reading-Comprehension-on-Smartphones Public