Aloe is an 'epistemic intelligence lab' seeking to build AI agents that understand their own limitations, reason critically, and self-improve. They are hiring for multiple engineering and research roles in Vancouver, focusing on AI that can be trusted with critical applications. Their vision involves democratizing AI access, accelerating discovery, defending against AI misuse, and respecting human cognition.
A starter prompt for Claude Code, what you'll need, and how to reach them.
You are an expert AI researcher and developer. The goal is to build a minimal proof-of-concept (MVP) of an 'epistemic AI agent' focused on self-awareness and uncertainty quantification. Your task is to develop a Next.js 16 App Router application with React 19, Tailwind v4, and AI SDK v6 with Gemini, running on Vercel with Neon Postgres. The MVP should be a web interface where a user can ask the agent a question. The agent, instead of just answering, should first: 1) State its confidence level in its answer (e.g., 'I am 85% confident in this answer because...'). 2) Identify any critical missing information or assumptions that would improve its answer. 3) Suggest a follow-up question or action the user could take to reduce uncertainty. The backend should handle the LLM interaction and basic state management. Focus on making the confidence and introspection clear and distinct from the raw answer. The build/verify gate is a working web application where I can input a question, and the agent responds with its confidence, identified limitations, and suggested next steps before providing its primary answer.
The audience interested in Aloe's advanced AI research and agent models would be prime candidates for the operator's agent testing and monitoring tools; highlight how your tools help build and evaluate the very agents Aloe is researching.
Aloe | Multiple Engineering & Research roles | ONSITE Vancouver, BC, Canada | Full-time Aloe is the epistemic intelligence lab building AI we can trust with humanity's most important applications. We work on research and products that bring introspection to AI and critical thinking to machines. Our Agent Models are designed for decision-making in the real world – AI that understands what it does and doesn't know, reasons about its limitations, and directs its own learning to self-improve. Please read the job description and include in your outreach why you fit not just the role but also our culture. We strive to be Level 5 humans, co-creating a deliberately developmental atmosphere. https://aloe.inc/careers The future we’re building for: - Commoditize access to expertise, capability, and AI’s economic promise for both businesses and end-users - Make new discoveries, products, and experiences possible at speeds unbound from human performance - Defend against negligent or malicious use of AI with internal governance and risk evaluation at the foundation - Invert the extraction economy with AI that respects human cognition instead of exploiting it
Standard LLM API access
Standard database for Next.js apps
Standard Next.js deployment platform
New: designing prompts for meta-cognition and self-reflection in LLMs - ~3-5 days of dedicated study and experimentation.
Learn it: Search getting-started ↗
Apply directly via the careers page: https://aloe.inc/careers
“I'm a solo operator deeply interested in epistemic AI and building trustworthy agents. I've developed a prototype demonstrating how an AI can articulate its confidence, identify knowledge gaps, and suggest clarification, which aligns directly with Aloe's mission. Happy to share a demo and discuss how my practical experience with AI agent development could contribute.”
Open the original ↗