Machine Learning Engineer 2

Twilio•,

40d•$128,560 - $170,000•Remote

About The Position

Twilio is hiring a Machine Learning Engineer for its Customer Memory work on the Conversation Memory team. The team builds the memory layer behind Twilio’s Conversational AI platform, extracting relevant information from conversations, resolving it to the correct customer, and retrieving it quickly when needed. This role builds and scales the pipelines that turn raw conversation data into structured, searchable customer memory, as well as evaluation systems that help keep memory accurate as the product evolves. Quality measurement is central to the work because incorrect, stale, or cross-customer memory can be worse than having no memory. The role is remote, but hiring is not available in San Francisco, Oakland, San Jose, or the surrounding areas. Occasional travel may be required for in-person project or team meetings. Twilio offers competitive pay, generous time off, parental and wellness leave, healthcare, a retirement savings program, and other benefits, with offerings varying by location. Compensation varies by location and may include participation in an equity plan and corporate bonus plan.

Requirements

Experience building backend services and demonstrated ability to learn new languages, frameworks, and tools as needed.
Ability and interest in investigating production system behavior and learning to operate services with rigor and accountability.
Experience building or working with evaluation or quality-measurement systems, such as regression detection or quality metrics, or desire to learn.
Solid grounding in observability practices, including structured logging, metrics, and debugging distributed systems in production.
Fluency with AI-assisted development tools, such as agents for code generation, review, and debugging, applied with deliberate direction and critical thinking.

Nice To Haves

Experience with Go and/or Python.
Experience with event-streaming systems such as Kafka.
Experience building LLM-backed products.
Practical experience applying machine learning, such as building, evaluating, or productionizing models.

Responsibilities

Build the extraction pipeline that transforms raw data into structured memory, tuning LLM prompts and schemas that determine what an AI agent remembers about a customer.
Own the retrieval path for finding relevant customer information in milliseconds, working with embeddings, vector search, and ranking.
Contribute to evaluation systems that measure memory quality through benchmarks and regression gates run before changes ship.
Diagnose production issues across ingestion, extraction, and indexing using logs, metrics, and traces.
Partner with Product and ML teams to translate customer problems into system changes and evaluation criteria.