- Design and build intuitive frontend interfaces for model understanding, such as:
- Docent, for analyzing long agent transcripts
- Monitor, for understanding internal model activations
- Implement scalable backend systems to power those interfaces
- Create APIs, SDKs, and protocols that interface between our products and external libraries
- Engage continuously with customers to gather requirements and feedback
- Maintain open source repositories and actively engage with the developer community
- Excels in rapid cycles of design, implementation, testing, and iteration
- Has strong design taste and a sharp attention to detail
- Writes well-designed code that enables team-wide collaboration
- Is proficient with our current stack: Python, Typescript, Next.js, AWS, serverless container runtimes
- Has experience engaging with customers and stakeholders
- Bonus: practical experience with language models (e.g., building apps on top of LLM APIs) and familiarity with fundamental ML concepts (e.g., ability to critically assess claims about model performance)
Skills Required
- Proficient with Python, Typescript, Next.js, AWS, and serverless container runtimes
- Experience engaging with customers and stakeholders
- Practical experience with language models and foundational ML concepts
What We Do
Transluce is an independent research lab that builds open, scalable technology for understanding AI systems and steering them in the public interest. Transluce means to shine light through something to reveal its structure. Today’s complex AI systems are difficult to understand—not even experts can reliably predict their behavior once deployed. Given AI's extraordinary consequences on society, we need scalable and open analyses of the capabilities and risks of AI systems. We are building open source, AI-driven tools to understand and analyze AI systems. We will apply these tools to open-weight models, so the world can vet our analyses and improve their reliability. Once our technology has been vetted, we will work with frontier AI labs and governments to ensure that internal assessments reach the same standards as our publicly vetted procedures. Email: [email protected]
