Backend Engineer — AI Agent Platform (Equity)

Sift Stack, Inc.; San Francisco (CA); On-site; USD 150,000 - 200,000; Full time

Job summary

Sift is hiring for a product engineering role focused on building AI agents for hardware telemetry. You will own features from MVP to production, interact with customers to shape the product, and design tool interfaces, execution environments, and scalable workloads.

Expect work spanning surface-level UI design to Kubernetes-based orchestration and real-time data processing. You will contribute to the MCP server, APIs, and evaluation suites, ensuring agents remain dependable, observable, and

Qualifications

• 3+ years of professional software engineering experience.
• Ability to own a product area and ship features end-to-end.
• Experience building APIs (REST, gRPC) and backend services.
• Working knowledge of distributed systems fundamentals.

Responsibilities

• Talk directly to customers and translate workflows into agent capabilities and dashboards.
• Design, ship, and operate systems that reason over large-scale time-series data and hardware context.
• Build tool interfaces, sandboxed execution, context/memory management, and guardrails for reliable agents.
• Develop and maintain MCP server and APIs powering agent capabilities.
• Create evaluation suites to measure agent effectiveness, latency, cost, and reliability.
• Work with Kubernetes-based deployment in cloud and on-prem environments.

Skills

Product ownership

Customer interviewing

APIs (REST, gRPC)

Distributed systems

Go

Tools

Go

Python

Rust

Kubernetes

Job description

Sift is hiring for a product engineering role focused on building AI agents for hardware telemetry. You will own features from MVP to production, interact with customers to shape the product, and design tool interfaces, execution environments, and scalable workloads.

Expect work spanning surface-level UI design to Kubernetes-based orchestration and real-time data processing. You will contribute to the MCP server, APIs, and evaluation suites, ensuring agents remain dependable, observable, and