T

TypeSafe AI

Develops system-one AI for software automation

Platform & Infra Engineer

Backend EngineeringUpdated on Nov 10
Full-timeSan Francisco, CA, USA$210k - $285k /yr

About the Job

Requirements

  • Have experience designing storage abstractions for scaled backend systems.
  • Have experience designing resilient systems and improving on-call experience on production systems.
  • Collaborate on technical and product design, advocate for needed requirements, and adjust specifications to changing requirements and feedback.
  • Have 5+ years of professional software engineering experience, including 3+ years of infrastructure or backend experience, on a team with rigorous engineering standards.
  • Have shipped backend services at scale with continuous delivery and uptime service-level agreements of at least 99.9%.
  • Have designed and operated a developer-facing application programming interface service.
  • Have experience with the capabilities and limitations of large language models from implementing them.

Responsibilities

  • Define application programming interfaces for a new AI inference experience and iterate on the design to make them useful and easy to use.
  • Create robust infrastructure for serving inference across multiple cloud providers.
  • Define abstractions to support future billing and business logic needs.
  • Expose new model features to users.
  • Build and maintain the application programming interface platform for inference, including user-facing, latency-sensitive services with uptime, reliability, and backward-compatibility requirements.

Desired Qualifications

  • Have previously built large-scale products or systems.

About the Company

TypeSafe AI

Develops system-one AI for software automation

TypeSafe AI builds machine-native automation inside software. Its flagship model, Jev, takes unstructured program state and returns typed, probabilistic decisions that software can act on, accessed via API in early access for workflows like routing, scoring, extraction, and verification. It uses a custom parallel sampler architecture and RL for Calibrated Decisions (RLCD) to control when and how decisions are made within applications. The company aims to embed programmable, reliable decision-making inside existing software to automate workflows, differentiating itself from consumer-focused chat AI and other AI vendors; it emerged from stealth in 2026 with a $40 million seed led by DCVC.

Company size
11-50
Company stage
Series A
Total funding
$910M
Headquarters
San Francisco, CA
Founded
2024

Recently Posted Jobs

Explore more opportunities at TypeSafe AI

See more jobs