AE Studio是一家位于洛杉矶的科技咨询公司，致力于通过应对即将到来的AGI（通用人工智能）未来，来增强人类的自主性。该公司正在招聘数据科学家和ML工程师，以支持其在AI对齐领域的工作。AE Studio专注于通过“被忽视的方法”来解决AI对齐问题，包括技术研究和政策制定。他们提供有竞争力的薪资和有意义的股权，鼓励远程或洛杉矶本地的员工。AE Studio的团队由顶尖的开发者、数据科学家、研究人员和创始人组成，他们将客户项目的利润再投资于有前景的AI对齐研究和雄心勃勃的内部项目。

🧠 AE Studio 专注于AI对齐工作，旨在确保AI的发展与人类的价值观相符。他们正在招聘数据科学家和ML工程师来支持这一目标。

💡 该公司采用“被忽视的方法”来解决AI对齐问题，这包括探索技术研究和政策制定。他们此前在脑机接口（BCI）领域验证了这种策略的有效性。

🛠️ AE Studio 提供的具体工作包括研究代理表示中的自我与他人重叠、使用稀疏自编码器进行特征引导、研究超出分布数据的的信息损失，以及与专注于对齐的初创公司合作等。

💰 AE Studio 提供有竞争力的薪资、有意义的股权，以及在内部研发项目和初创企业中的股权。他们还提供一个重视自主性、长期思考和实际影响的文化。

Published on April 21, 2025 8:35 PM GMT

AE Studio is Hiring: Data Scientists & ML Engineers for AI Alignment Work

Location: Remote or Los Angeles (in-person strongly encouraged)
Type: Full-time
Compensation: Competitive salary + meaningful equity in client and Skunkworks ventures

Who We Are

AE Studio is an LA-based tech consultancy focused on increasing human agency, primarily by making the imminent AGI future go well. Our team consists of the best developers, data scientists, researchers, and founders. We do all sorts of projects, always of the quality that makes our clients sing our praises. 

We reinvest those client work profits into our promising research on AI alignment and our ambitious internal skunkworks projects. We previously sold one of our skunkworks for some number of millions of dollars.

We have made a name for ourselves in cutting-edge brain computer interface (BCI) R&D, and after working on this for the past two years, we have made a name for ourselves in research and policy efforts on AI alignment. We want to optimize for human agency, if you feel similarly, please apply to support our efforts.

What We’re Doing in Alignment

We’re applying our "neglected approaches" strategy—previously validated in BCI—to AI alignment. This means backing underexplored but promising ideas in both technical research and policy. Some examples:

Investigating self-other overlap in agent representationsConducting feature steering using Sparse Autoencoders Looking into information loss with out of distribution data Working with alignment-focused startups (e.g., Goodfire AI)Exploring policy interventions, whistleblower protections, and community health

You may have read some of our work here before but for a refresher, feel free to go to our LessWrong profile and get caught up on our thought pieces and research. Previous posts have been met with great reception by the likes of Eliezer Yudkowsky and Emmett Shear, so we’re up to something good. 

Interested in more information about what we’re up to? See a summary of our work here: https://ae.studio/ai-alignment 

About You

Passionate about AI alignment and optimistic about humanity’s future with AIExperienced in data science and ML, especially with deep learning (CV, NLP, or LLMs)Fluent in Python and familiar with calling model APIs (REST or client libs)Love using AI to automate everything and move fast like a startupProven ability to run projects end-to-end and break down complex problemsComfortable working autonomously and explaining technical ideas clearly to any audienceFull-time availability (side projects welcome—especially if they empower people)Growth mindset and excited to learn fast and build cool stuff

Bonus Points

Side hustles in AI/agency? Show us!Software engineering chops (best practices, agile, JS/Node.js)Startup or client-facing experienceBased in LA (come hang at our awesome office!)

What We Offer

A profitable business model that funds long-term researchFull-time alignment research opportunities between client projectsEquity in internal R&D projects and startups we help launchA team of curious, principled, and technically strong peopleA culture that values agency, long-term thinking, and actual impact

AE employees who stick around tend to do well. We think long-term, and we’re looking for people who do the same.

How to Apply

Apply here: https://grnh.se/de858ba44us 

Discuss