Ingrid Wu — Product leader & developer
Ingrid Wu is a New York–based product leader and developer focused on open source, NLP, and building "agentic" AI workflows that help teams systematically improve AI products. Her site showcases projects, technical notes, and practical write-ups on AI product management and tooling.
Key features
- Agentic workflows: practical approaches to orchestrating AI components and agents for end-to-end automation.
- NLP and LLM-focused work: experience applying NLP models and techniques to real products and metadata generation.
- Product + engineering blend: shipping features, improving data pipelines, and implementing scalable ML infra.
- Open-source and prototyping: early-stage projects and demos (Sellr, Nexa) demonstrating rapid iteration.
- Practical optimization: metadata/SEO optimization for marketplaces and user-facing product improvements.
Primary technical components and tools referenced
- Cloud and serverless deployment (AWS Lambda) for production serverless endpoints.
- Machine learning tooling: TensorFlow for model building (e.g., SVD models) and model inference.
- Data integrations: NASDAQ data ingestion, marketplace metadata pipelines, and SEO-focused content generation.
- Web and product stack: small web demos and prototypes (Vercel-hosted frontends).
Use cases
- Automating content generation for marketplace listings to reduce manual entry and improve discoverability.
- Rapid prototyping of voice- or agent-driven experiences (e.g., Alexa skill for market insights).
- Informing product teams on integrating LLMs/NLP into feature roadmaps and production systems.
- Experimenting with SEO and metadata strategies for content and music discovery platforms.
Target users
- Product managers and founders building AI-first products.
- Engineers and ML practitioners looking for practical examples of NLP/agentic workflows.
- Teams seeking to automate content and metadata generation to scale marketplace operations.
Unique selling points
- Blend of product leadership and hands-on engineering experience.
- Focus on practical, deployable solutions rather than purely academic models.
- Concrete project examples showing end-to-end implementation and iteration.
