I’m building a predictive analytics solution that classifies stock-market patterns, and I need an AI/ML engineer with a proven track record (five years or more in production-level work) to own the entire model-development cycle. The core of the job is to design, train, and validate a classification model that ingests historical and live market feeds, then outputs actionable signals. You’ll work in Python, choose and fine-tune the most appropriate machine-learning or deep-learning architecture, and put in place a data pipeline that can scale and be optimised for latency and accuracy. You should already be comfortable with feature engineering on financial time series, rigorous back-testing, hyperparameter tuning, and deploying models to a cloud environment or an on-prem server—whichever we agree best suits production. Clean, well-documented code is non-negotiable, as I plan to integrate the model into an existing trading stack. Deliverables: • End-to-end classification model with reproducible training scripts • Automated data pipeline and evaluation metrics dashboard • Brief technical report summarising methodology, performance, and next-step considerations If you’ve shipped similar stock-market ML solutions, please share links or repo snippets alongside a short note on the techniques and libraries you used. I’ll disclose the full project scope and datasets once I shortlist candidates.