Juan P.Private profile: Upwork shows its earnings only to signed-in members, so we check it every few weeks, not daily. MRR here is a rough estimate (Upwork's six-month earnings ÷ months active). Last read Nov 5, 2026.

I help data-heavy teams turn messy data into confident product decisions—using causal inference, machine learning, behavioral analytics, and rigorous AI evaluation. 𝗪𝗛𝗔𝗧 𝗜 𝗖𝗔𝗡 𝗛𝗘𝗟𝗣 𝗪𝗜𝗧𝗛 - 🧪 𝗠𝗲𝗮𝘀𝘂𝗿𝗲 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗶𝗺𝗽𝗮𝗰𝘁: Design experiments and causal analyses that clarify whether a product, intervention, or feature actually changed outcomes. - 🧠 𝗕𝘂𝗶𝗹𝗱 𝗠𝗟 𝗮𝗻𝗱 𝗔𝗜 𝘀𝘆𝘀𝘁𝗲𝗺𝘀: Develop predictive models, segmentation approaches, and practical LLM/RAG workflows for real product questions. - 🔎 𝗘𝘃𝗮𝗹𝘂𝗮𝘁𝗲 𝗔𝗜 𝘀𝘆𝘀𝘁𝗲𝗺𝘀: Compare models, prompts, retrieval strategies, and human judgment so LLM systems improve with evidence rather than guesswork. I work end to end: clarifying the decision, pulling and cleaning real-world data, building the right model or analysis, and communicating results in a form technical and non-technical teams can act on. My background combines PhD-level training in educational data science with peer-reviewed research in interpretable AI, learner modeling, and equitable educational technology. I have particular depth in EdTech and learning-science problems, but the underlying toolkit applies to any product team working with complex behavioral or longitudinal data. 𝗦𝗲𝗹𝗲𝗰𝘁𝗲𝗱 𝘁𝗼𝗼𝗹𝘀: Python, SQL, PySpark, R, scikit-learn, PyTorch, TensorFlow/Keras, Databricks, AWS, BigQuery, LangChain, and LangGraph. You can expect proactive communication, honest scoping, documented assumptions, reproducible work, and early flags when a question needs reframing.