Platform Engineer

Engram Lab•San Francisco, CA

4d•Onsite

About The Position

Engram's mission is to create AI that deeply understands user context and history, moving beyond generic models to personalized 'engrams'—compact memories that capture individual knowledge. This approach enables AI to scale by learning from user context at training time for better inference. Engram is already collaborating with major AI players like Microsoft, Notion, and Harvey, and has secured significant funding from leading venture capital firms. The company is seeking to build AI that learns about the user's world, not just the general world. This role is for one of Engram's first software engineers, joining a team of machine learning researchers and performance engineers. The focus will be on building the personalization and continual learning API, which powers models and agents that learn from user context. Specifically, the role involves architecting and implementing the data infrastructure, including designing secure data storage for user context and model weights, developing performant systems for data ingestion and preprocessing for continual training, building data observability for quality control and reproducibility, and creating reliable integrations with customer APIs and other data sources. The engineer will be a key architect, working closely with researchers, performance engineers, and customers. This is a founding hire position with significant influence on engineering practices, code review, testing, on-call responsibilities, security posture, and team building. The role is based in Engram's San Francisco office and requires an in-person presence.

Requirements

5+ years of software engineering experience, building and scaling systems in a high-stakes environment
Experience building APIs or infrastructure that other developers consume: SDKs, multi-tenant platforms, developer tooling
Taken a successful product or open-source project from zero to one
Product-driven and enjoy working across the stack
Worked somewhere with a rigorous engineering culture, where security and reliability are top priorities

Nice To Haves

Background working with ML teams and familiarity with the modern ML serving and training stack
Prior early-stage experience

Responsibilities

Designing a secure data storage architecture for raw user context and per-user model weights
Developing performant systems for ingesting and preprocessing data on a regular basis for continual training
Building data observability for quality control and reproducibility
Reliable integrations with customer APIs and other data sources
Setting the bar for engineering at Engram: code review, testing, on-call, and a security posture
Helping build the engineering team around you and influencing our engineering culture as we scale