Member of Technical Staff, Agent Harness

About Hark

Hark is an artificial intelligence company building advanced, personalized intelligence. One that is proactive, multimodal, and capable of interacting with the world through speech, text, vision, and persistent memory.

We're pairing that intelligence with next-generation hardware to create a universal interface between humans and machines. While today's AI largely operates through chat boxes and decade-old devices, Hark is focused on what comes next: agentic systems that interact naturally with people and the real world.

To get there, we're developing multimodal models and next-generation AI hardware together - designed from the ground up as a single, unified interface for a new era of intelligent systems.

About the Role

You'll build the backend systems that make Hark's AI agent actually work — reliable, fast, and production-grade.

That means the hard infrastructure problems: high-concurrency services, low-latency streaming, state management for long-running agent workflows, and the execution layer that connects model outputs to real-world actions. You are building the nervous system of an agentic product.

This is a high-ownership role on a small team. You'll work directly with model researchers and platform engineers, and the systems you build will determine whether Hark feels like a slow chatbot or something genuinely new.

Responsibilities

• Core Runtime Architecture: Design and build the high-concurrency backend services responsible for agent orchestration, tool execution, and real-time streaming.
• Systems Reliability: Develop robust failure-handling, retry logic, and state management for long-running agentic workflows.
• Platform Primitives: Architect the foundational APIs and services—including memory retrieval systems and sandboxed execution environments—that power the entire Hark ecosystem.
• Performance Engineering: Optimize the stack for low-latency streaming and high-throughput data processing to ensure seamless agent-user interactions.
• Full-Cycle Ownership: Lead features from low-level design and prototyping to production deployment, monitoring, and performance tuning.
• System Observability: Implement deep instrumentation and automated evaluation frameworks to track system health and model quality regressions.

Requirements

• Backend & Systems Mastery: 5+ years of experience building mission-critical backend systems. You are an expert in concurrency, networking protocols, and distributed systems.
• Production at Scale: Proven track record of shipping APIs and infrastructure that handle real-world traffic, with a deep understanding of horizontal scaling and "day 2" operations.
• AI System Intuition: Experience integrating LLMs into backend pipelines. You understand the unique failure modes of non-deterministic systems and how to wrap them in deterministic, reliable code.
• Language Proficiency: Strong experience in at least one systems-level language (e.g., Go, Rust, C++, or Java).
• Infrastructure Mindset: Comfort with cloud-native architectures (AWS/GCP), container orchestration, and building secure, isolated execution sandboxes.
• Technical Communication: Ability to articulate complex architectural tradeoffs and collaborate with model researchers to bridge the gap between AI and production-grade software.

Bonus Qualifications

• Hands-on experience with gRPC, WebSockets for high-performance streaming.
• Prior work with vector databases, distributed caching, or custom memory management systems.
• Deep knowledge of Kubernetes, microservices security, or serverless execution patterns.
• Experience building developer-facing APIs or SDKs.

Compensation

The US base salary range for this full-time position is between $170,000 - $400,000 annually.

The pay offered for this position may vary based on several individual factors, including job-related knowledge, skills, and experience. The total compensation package may also include additional components and benefits depending on the specific role. This information will be shared if an employment offer is extended.