AI Engineer Internship

Skills this role screens for

The work

About the Role

We are looking for an AI Engineer Intern to build intelligent systems, AI-powered workflows, and scalable data infrastructure that power next-generation AI products.

You will work on integrating and evaluating Large Language Models (LLMs), building automated data pipelines, developing AI-powered information extraction systems, and creating intelligent search and discovery capabilities. Depending on the product requirements, your work may involve model APIs, embeddings, retrieval systems, web crawling, structured data extraction, and AI agents.

Inspired by modern AI infrastructure platforms such as OpenRouter, you will gain practical experience working with multiple AI models, designing reliable LLM workflows, measuring model performance, and building systems that move beyond experimental notebooks into production applications.

This role is ideal for someone who enjoys Python, applied AI, backend engineering, and solving complex technical problems with measurable outcomes.

The work

Responsibilities

• Build AI-powered pipelines for data acquisition, processing, enrichment, and automation.
• Integrate and experiment with LLMs from providers such as OpenAI, Anthropic, Google, and OpenRouter.
• Develop workflows for structured extraction, classification, summarization, tagging, and information synthesis.
• Build scalable web crawlers and data ingestion pipelines to collect information from websites, APIs, documentation, GitHub, and other public sources.
• Develop systems to discover, organize, and enrich information about AI models, tools, APIs, datasets, research papers, benchmarks, and emerging technologies.
• Implement embedding generation, semantic search, vector retrieval, and Retrieval-Augmented Generation (RAG).
• Build and maintain data processing pipelines, validation rules, deduplication, and entity resolution.
• Experiment with model selection, prompt engineering, structured outputs, and tool calling.
• Design evaluation pipelines to measure model accuracy, relevance, latency, reliability, and cost.
• Implement robust error handling, retries, rate-limit management, caching, and fallback strategies for AI API integrations.
• Develop backend services and APIs that expose AI-powered capabilities to frontend applications.
• Monitor pipeline quality, investigate failures, and improve data freshness and system reliability.
• Research new models, AI frameworks, and infrastructure tools to identify practical improvements.
• Document experiments, technical decisions, and performance benchmarks.

Who they want

Requirements

• Strong proficiency in Python and solid programming fundamentals.
• Understanding of APIs, JSON, HTTP, asynchronous programming, and backend development.
• Hands-on experience building at least one meaningful project using an LLM API or open-source language model.
• Understanding of LLM fundamentals, prompt engineering, tokens, context windows, and structured outputs.
• Familiarity with data processing, web scraping, and data ingestion using tools such as Requests, BeautifulSoup, Playwright, or Scrapy.
• Experience with PostgreSQL, SQLite, MongoDB, or another database.
• Familiarity with Git and GitHub.
• Understanding of data structures, algorithms, debugging, and writing maintainable code.
• Ability to evaluate technical approaches through experiments rather than relying solely on model outputs or assumptions.
• Strong problem-solving skills and the ability to work independently in a fast-paced startup environment.
• A portfolio of AI projects, GitHub repositories, research implementations, or working demos that demonstrate practical engineering ability.

Nice to have

Good to Have

Who Should Apply?

We want engineers who enjoy building and experimenting with real AI systems, not just creating chatbot demos. You should be comfortable reading documentation, testing multiple approaches, measuring results, and turning promising experiments into reliable software.

Strong Python fundamentals, practical project experience, and the ability to explain your technical decisions are more important than having used every framework listed above.

• Experience with OpenRouter, OpenAI, Anthropic, Gemini, or other LLM APIs.
• Familiarity with LangChain, LlamaIndex, Hugging Face Transformers, or agent frameworks.
• Experience with embeddings, vector databases such as Qdrant, pgvector, or Pinecone, and RAG pipelines.
• Knowledge of model evaluation, LLM observability, tracing, and experiment tracking.
• Familiarity with MCP, function calling, multi-agent systems, or tool-use workflows.
• Experience with Docker, cloud deployment, Redis, task queues, or workflow orchestration.
• Understanding of model routing, provider fallbacks, latency optimization, token usage, and inference cost optimization.
• Experience with PyTorch, fine-tuning, open-source models, or machine learning fundamentals.