Machine Learning Engineer

WireScreen is a fast-growing Series A startup bringing clarity to one of the world’s most complex business landscapes. We’re building the go-to intelligence platform for navigating global supply chains and China-related risk—revealing the networks, relationships, and financial ties behind companies so our customers can see connections, remove obstacles, and make confident decisions. Backed by Sequoia Capital and Harpoon Ventures, our team includes a two-time Pulitzer Prize–winning journalist and senior engineers from Google, Twitter, and Oracle. We launched just three years ago and already have strong traction with top-tier government customers—and we’re just getting started. If you’re excited to make global systems more transparent and secure, now’s the perfect time to join us. ABOUT THE ROLE: As a Machine Learning Engineer at WireScreen, you will be working across our data systems to unlock the source of truth behind one of the world’s economic powerhouses: China. This role is critical to building models that support the two fundamental pieces of our business: entity resolution and our knowledge graph. These systems are the cornerstones that uncover hidden connections within the Chinese economy. You will be contributing to our existing MCP servers and agentic pipelines to classify, cluster and enrich across millions of entities. You will be working with engineering, product and research to expand our AI toolkit, scaling our data ingestion processes, building analytics and insights into our platform at a scale not previously achievable. Reporting directly to the VP of Engineering, you will work with our data team, enrichment team and product to establish our source of truth and expand our machine learning capabilities. WHAT YOU'LL DO: - Fine tune our existing entity resolution algorithms to uncover hidden connections between people and organizations across China - Expand our knowledge graph with alternative data to map out the power structure of China - Train, test and deploy ML models that operate on tens of millions of records daily - Work with Product to define and implement evaluation harnesses for classical ML and agentic systems - Build agent workflows into internal tools to improve the scale and speed of our Research team WHAT WE’RE LOOKING FOR: - 4+ years of experience working on clustering-type ML problems, ideally in the domain of knowledge graphs / entity resolution, but other domains could include; recommendation engines, cohort analysis, outlier/anomaly detection - End-to-end machine learning model experience in production; that you’ve stood up a service including experimenting, training, testing and tuning a job against a dataset all the way through to deployment and beyond. Model families could include clustering, classification/regression, dimensionality reduction and embeddings, nearest-neighbor/similarity methods (e.g. KNN, SVM), ensembles, NLP, and deep learning. - Significant experience with python programming and SQL NICE TO HAVE: - Experience working with Frontier/SOTA models and/or fine-tuning your own LLMs for specific tasks - Working on problems across large, heterogeneous, messy unstructured datasets and/or with semantic search, computer vision (especially OCR), or linear optimization problems - Experience with any of the following technologies: PySpark, Temporal, FastAPI, Scikit-learn, NumPy, Docker, Terraform, Kubernetes - Early-stage startup experience (Series B or earlier) - B2B SaaS experience #LI-LG1 WHAT YOU'LL LOVE ABOUT WIRESCREEN At WireScreen, you'll do high-impact work that helps shape the global economy and power competition. We’re a mission-driven team with a growth mindset—curious, collaborative, and unafraid to take on bold challenges. You’ll be empowered to act, heard when you speak, and supported as you grow. With strong market momentum and ambitious goals, this is an exciting time to join us and help build something that truly matters. BENEFITS & PERKS At WireScreen, we care deepl…