AI Engineer & LLM Tooling FALL 2026 Upaid Internship

Public source summary from intern-list.com / Jobright.

Carboncopies Foundation is an international nonprofit organization dedicated to advancing the science and technology of whole brain emulation. The AI Engineer Intern will build, integrate, and maintain LLM-powered applications, agent pipelines, RAG infrastructure, developer tooling, and full-stack AI features that support the organization’s research workflows.

LLM Application Development

Build and ship LLM-powered features end to end, from RAG pipelines and embeddings/semantic search to streaming chat experiences (SSE) and production FastAPI/Node.js APIs, designed for token efficiency

AI Tooling & Agentic Workflows

Drive adoption of AI-assisted engineering with tools like Claude Code, Antigravity CLI, and the Mistral API family. Build internal CLI tools, prompts, and automation that accelerate the whole engineering team

RAG & Vector Infrastructure

Design and maintain retrieval-augmented generation pipelines with vector databases (ChromaDB, Pinecone, or Qdrant), including embedding generation, chunking strategies, and re-ranking to keep prompts lean Model Integration & Cost Optimization Integrate and benchmark multiple AI providers (Mistral, OpenAI, OpenRouter) with real-time token tracking, spend monitoring, and automatic routing to the cheapest capable model.

As a nonprofit, we prioritize free tiers, credits, and open-weight alternatives wherever possible CI/CD & Developer Experience Automate build, test, and deployment pipelines with GitHub Actions and Docker, keeping research code flowing smoothly into production Full-Stack Development Contribute across the stack (Next.js/React/SolidJS frontends, PostgreSQL backends) to build and ship user-facing AI tools like our CarbonGPT research assistant

Qualifications: Strong TypeScript and Python skills, with experience building production APIs and asynchronous backend services Hands-on work with LLM APIs (Mistral, OpenAI, OpenRouter), embeddings, streaming responses, and RAG pipelines Practical experience with agentic coding assistants such as Claude Code, Antigravity CLI, Cursor, or similar Comfortable across Next.js, React, Node.js, and relational databases (PostgreSQL, Drizzle ORM, or similar) Proficiency with Git, Docker, and GitHub Actions CI/CD Ability to take an AI feature from idea to live production, including auth, and monitoring Currently pursuing (or holding) a degree in Computer Science, Software Engineering, or a related technical field Bonus points for experience with prompt caching, context-window management, and other token-efficiency techniques You think in cost per request.

You know how to exploit free tiers, nonprofit credits, and open-source/self-hosted models (e.g., Ollama, vLLM) to keep spend near zero without sacrificing quality You should know how to prompt, configure, and orchestrate these tools to multiply your output Cloud experience (AWS) is a plus

Benefits: Remote work arrangement Flexible time commitment, approximately 20 hours per week