Jie Wei

• • Delivered production-grade AI and backend systems for multiple clients across healthcare, SaaS, and automation, supporting scalable and secure deployments

• • Designed and implemented LLM-based applications using LangChain, enabling retrieval-augmented generation (RAG), structured outputs, and intelligent decision workflows

• • Built and maintained backend services using Python (FastAPI) and Node.js, supporting high-throughput APIs, async processing, and third-party integrations

• • Developed and optimized machine learning models using PyTorch and TensorFlow, improving inference performance and reliability in production environments

• • Implemented NLP and Voice AI pipelines, including speech-to-text, intent classification, and conversational logic for real-time applications

• • Collaborated directly with founders, product managers, and stakeholders to translate business requirements into scalable, maintainable technical solutions 2023

• • Enhanced backend workflows for AI-driven document processing in Python, improving throughput and extraction quality for legal and medical use cases.

• • Integrated image and text inference services into APIs, including OCR-based pipelines and NLP post-processing for structured outputs.

• • Deployed and operated cloud-native components on Azure with secure data handling, auditability, and access controls aligned with best practices.

• • Partnered with product and data teams to define quality metrics, edge cases, and automated checks for model and pipeline reliability.

• • Produced API references and runbooks that reduced onboarding time and improved support and maintenance.

• Enhanced backend workflows for AI-driven document processing in Python,

improving throughput and extraction quality for legal and medical use cases. • Integrated image and text inference services into APIs, including OCR-based pipelines

and NLP post-processing for structured outputs. • Deployed and operated cloud-native components on Azure with secure data

handling, auditability, and access controls aligned with best practices.

• Partnered with product and data teams to define quality metrics, edge cases, and

automated checks for model and pipeline reliability. • Produced API references and runbooks that reduced onboarding time and improved

support and maintenance.

• • Integrated predictive models into enterprise applications using FastAPI and PostgreSQL, delivering low-latency scoring APIs and batch jobs.

• • Collaborated with data scientists to productionize ML workflows and improve monitoring for drift, failure modes, and data quality issues.

• • Built internal tooling and dashboards to streamline review and troubleshooting of model outputs and pipeline runs.

• • Used Docker and Azure Pipelines to standardize builds and deployments across environments, improving release safety and consistency.

• • Authored design docs and system architecture notes to align stakeholders and speed up delivery.