Sentient Futures
Back to mentors
Allen Lu
AI tools to empower advocates AI’s nearterm impact on animals Post-AGI transition/Transformative AI Longtermism and future suffering Macrostrategy, philosophy, and cause prioritisation

Allen Lu

Technical Strategies & Interventions for Multi-species AI Safety

AI Product Engineer (Independent)

Bio

Hi, I'm Allen! I'm an interdisciplinary product engineer engaged at the intersection of AI safety & animal welfare. My work mainly involves developing (mostly technical) solutions to reduce animal suffering, especially from AI-driven harms. I've facilitated fellowships for NYU, Sentient Futures, and Electric Sheep on topics around EA & AIxAnimals. I lead the THL policy team in NYC, pushing for legislative action for farmed animals. I also write "Bytes of Compassion", a Substack publication exploring AI safety, strategy, and animal advocacy.

Mentee must-haves/nice-to-haves

General requirements:

  • Genuine interest in both AI safety and animal welfare
  • Self-starter mindset, and able to make measurable progress between meetings
  • Bonus: previous work in animal advocacy, AI safety, EA or adjacent fields (please note below!)

The below skills are NOT hard requirements, though having experience in these is a huge plus. If you don’t have any of these but are still genuinely interested in the project, I’d encourage you to apply! I’m happy to work with you and I’m sure we’ll both learn a lot in the process. Project 1 (Technical Program Design):

  • Background in software engineering, computer science, and other ML/AI concepts
  • Experience in educational programs, course design, facilitation, or something similar
  • Strong communication skills, and comfortable outreaching individuals/orgs to create key partnerships to contribute to program design & deployment Project 2 (Research Engineering - Benchmarks & Evals):
  • Background in software engineering, CS, research, data, etc.
  • Experience with Python, Git, or other coding languages
  • Familiarity with LLM evaluation frameworks or benchmarking
  • General knowledge in AI safety/alignment concepts and research methodology

Mentee role

You decide! I’m flexible, and happy to work in a way that works for you. Here are some options I see:

  • Co-builders: Take ownership of major components and drive them to completion
  • Contributors: Execute specific tasks like curriculum development, research, or content creation with guidance
  • Supporters: Assist with tasks, such as research, logistics, and implementation

For sample mentee tasks, see the project description above.

Mentor support

Happy to help with generally brainstorming project ideas, being a sound board to bounce ideas off of, connecting you with others in my network that may be helpful to provide insights, etc. For technical projects, I can review your code, program with you side-by-side, suggest tools/APIs for usage, potentially provide credits or other resources, etc.

Questions for applicants

General questions (for everyone):*

  1. Which project interests you most? Why do you want to work on this project? (50 words)
  2. What's your background? Please briefly describe your experience and any relevant skills related to this project (technical, research, advocacy, etc.). (100 words)
  3. What draws you to work on AI×Animals-related projects? How have you engaged with this field so far (i.e. read, built, or explored anything in this area)? Totally ok if you haven’t, but just so I know! (100 words) *if you’ve already answered these questions in a different part of the application, feel free to just copy and paste your answers

Custom Project Question: What's your project idea in 2-3 sentences? What would success look like after 3 months? What concrete outputs or outcomes would demonstrate progress? (100 words)

Mentor-led project

Technical Strategies & Interventions for Multi-species AI Safety

I’m exploring the creation of a nonprofit, and have a couple ideas for projects listed below. In your application, feel free to specify if there’s one you’d like to work on, or propose a new idea entirely! (does not have to be related to my stuff at all and can be your own project)

Project 1: Technical Program Design Help design a professional course, fellowship, or program geared towards animal advocates and junior engineers, upskilling technical skills to be able to effectively contribute to AIxAnimals work (e.g. working on frontier models). This is modeled after other programs by Bluedot, Electric Sheep, and Sentient Futures. Project work may include:

  • Researching other courses in AI safety, animal welfare, research, etc. and compiling reports on the current landscape
  • Contributing to general program outline - syllabus, readings, course structure, etc.
  • Creating technical assignments (computer science, ML/AI, etc.) within program itself
  • Facilitating pilot workshop to participants with documented feedback and iteration plan

Project 2: Research Engineering - Benchmarks & Evals Contribute to benchmarking tools and other eval methods to measure how frontier AI models reason about non-human beings, inspired by work from AnimalHarmBench 2.0. *Important* I’m also mentoring in the SPAR program and will be taking on additional mentees for this project through there. As such, the start date for this project may be slightly different from the Sentient Futures-appointed date to coordinate multiple mentees for this project (still mid-feb for sure though) Project work may include:

  • Contributing to general benchmark code (Python scripts, documentation)
  • Develop technical documentation explaining benchmark design, scoring methodology, etc.
  • Doing research and co-authoring sections to a research paper draft or blog post on findings

For more detailed information, feel free to check out the project doc here.