Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Machine Learning Engineer with 8+ years of experience building production AI systems, LLM-powered workflows, developer automation, and scalable machine learning infrastructure. Experienced in Python and TypeScript/Node.js for tool-calling and prompt orchestration, retrieval-augmented generation, and evaluation frameworks. Focused on reliability through permission controls, auditability, testing, observability, and rollback procedures, including human-in-the-loop approval workflows for production changes.…Machine Learning Engineer with 8+ years of experience building production AI systems, LLM-powered workflows, developer automation, and scalable machine learning infrastructure. Experienced in Python and TypeScript/Node.js for tool-calling and prompt orchestration, retrieval-augmented generation, and evaluation frameworks. Focused on reliability through permission controls, auditability, testing, observability, and rollback procedures, including human-in-the-loop approval workflows for production changes.

Jiayong Lin

AI Engineer, Data Scientist, Software Engineer, +1





Machine Learning Engineer with 8+ years of experience building production AI systems, LLM-powered workflows, developer automation, and scalable machine learning infrastructure. Experienced in Python and TypeScript/Node.js for tool-calling and prompt orchestration, retrieval-augmented generation, and evaluation frameworks. Focused on reliability through permission controls, auditability, testing, observability, and rollback procedures, including human-in-the-loop approval workflows for production changes.…Machine Learning Engineer with 8+ years of experience building production AI systems, LLM-powered workflows, developer automation, and scalable machine learning infrastructure. Experienced in Python and TypeScript/Node.js for tool-calling and prompt orchestration, retrieval-augmented generation, and evaluation frameworks. Focused on reliability through permission controls, auditability, testing, observability, and rollback procedures, including human-in-the-loop approval workflows for production changes.

Available to hire

Machine Learning Engineer with 8+ years of experience building production AI systems, LLM-powered workflows, developer automation, and scalable machine learning infrastructure.

Experienced in Python and TypeScript/Node.js for tool-calling and prompt orchestration, retrieval-augmented generation, and evaluation frameworks. Focused on reliability through permission controls, auditability, testing, observability, and rollback procedures, including human-in-the-loop approval workflows for production changes.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Language

English

Intermediate

Work Experience

Machine Learning Engineer at Meta

July 1, 2022 - Present

Designed LLM-based engineering agents that decompose technical requests, retrieve codebase context, invoke internal tools, and generate proposed code, tests, documentation, and pull-request summaries for engineer review. Built codebase-aware retrieval pipelines using embeddings, semantic search, and repository metadata. Implemented prompt orchestration and tool-calling workflows in Python and TypeScript with safe interactions to Git, issue tracking, test environments, documentation systems, and deployment services. Developed human-in-the-loop review flows with role-based permissions, approval states, sensitive-path restrictions, audit logs, and rollback procedures before production-facing changes. Created evaluation/benchmarking suites for patch correctness, retrieval relevance, tool-selection accuracy, hallucination rates, reviewer acceptance, latency, and task quality; integrated static analysis and CI validation into PR workflows. Built release agents for summarizing merged changes,

Machine Learning Engineer at The Michaels Companies, Inc

August 1, 2020 - July 31, 2022

Built Python and Node.js automation services connecting Jira, Git repositories, CI pipelines, internal documentation, and model-deployment workflows to reduce manual coordination across engineering and analytics. Developed an AI-assisted documentation workflow that indexes technical specs, experiment results, data definitions, and runbooks, then uses semantic retrieval to answer and draft implementation notes. Designed recommendation and personalization systems using customer behavior, product metadata, seasonality, and marketing signals. Implemented automated model-validation gates in GitHub Actions and Jenkins including data-quality checks, unit tests, offline evaluation, regression detection, and approval requirements before deployment. Created reusable utilities to test and benchmark recommendation quality, segmentation stability, inference latency, data drift, and experiment performance. Established model versioning, release documentation, monitoring, audit trails, and rollback fo

Data Scientist at University of Maryland, Baltimore County

February 1, 2018 - May 31, 2020

Developed NLP and information-retrieval systems to classify technical documents, extract entities and relationships, and surface relevant research context via semantic and keyword search. Built reproducible Python workflows for supervised learning, anomaly detection, document analysis, feature engineering, model evaluation, and automated experiment reporting. Created benchmarking suites comparing model accuracy, retrieval quality, runtime, robustness, and error patterns across academic, cybersecurity, biomedical, and institutional datasets. Built REST-based research services and automated data-processing pipelines using Python, SQL, TensorFlow, scikit-learn, and Pandas/NumPy. Applied explainability and structured error analysis to communicate model risk and limitations to stakeholders. Mentored undergraduate researchers and maintained technical documentation, experiment records, source control practices, and reproducible testing standards.

Research Assistant at University of Maryland, Baltimore County

September 1, 2016 - February 28, 2018

Prototyped machine learning and rule-based systems for cybersecurity anomaly detection, document classification, and knowledge-driven question answering. Built knowledge graphs using RDF and SPARQL to connect technical documents, entities, relationships, and evidence for contextual retrieval. Automated data collection, preprocessing, validation, and experiment execution using Python, SQL, Git, and reproducible testing workflows. Evaluated supervised models with cross-validation, error analysis, performance benchmarks, and documented acceptance criteria. Collaborated with faculty and graduate researchers to translate requirements into software prototypes, APIs, documentation, and presentations.