Machine Learning Projects (Real AI Systems, Real Portfolio)
The AI projects that impress are real systems, not another model-from-scratch notebook. These are real AI/ML systems you fix in a live cloud workspace - a RAG pipeline, an eval harness, chunking that wrecks retrieval - each one a portfolio piece.
Why these beat a from-scratch notebook
Most AI jobs are building reliable systems around models - retrieval, evaluation, cost, integration. These practice that engineering, which is what teams actually hire for.
18 projects to start with
Build an End-to-End RAG PipelineAI/ML · SeniorBuild an LLM Evaluation Harness (Golden Set)AI/ML · SeniorRoute LLM Requests to Cheaper ModelsAI/ML · SeniorBuild a Tool-Calling Agent LoopAI/ML · SeniorStream a RAG Answer With Inline CitationsAI/ML · SeniorBuild an LLM Content Moderation ServiceAI/ML · MidCut OpenAI Costs by Caching With RedisAI/ML · MidPower RAG Search With a Vector DatabaseAI/ML · MidFix the Chunking That's Wrecking RAG RetrievalAI/ML · MidRe-Rank RAG Results So the Best Answer Lands on TopAI/ML · MidMake an LLM Pick the Right Tool (Function Calling)AI/ML · MidRestore a Broken LLM API IntegrationAI/ML · JuniorHarden an LLM Pipeline Against API FailuresAI/ML · JuniorParse JSON From an LLM (Strip Markdown Fences)AI/ML · JuniorBuild a Text Summarization Endpoint With an LLMAI/ML · JuniorAdd Conversation Memory to a ChatbotAI/ML · JuniorRestore Token Streaming on a Chat EndpointAI/ML · JuniorCache LLM Responses in RedisAI/ML · Junior
Pick by the AI skill to show
Want to show RAG skill? The retrieval and chunking ones. Reliability? The eval and routing ones. Each finished fix is evidence on your portfolio.
Build your portfolio free. You are in a real cloud workspace in 30 seconds - fix real systems, show real work.
Start free →