Agent Infrastructure Engineer – Core Harness, Superagent

This is a fully remote position, open to applicants in India.

📋 Description

• Take charge of the architecture, development, and progression of Superagent, the essential agent harness that powers conversations, tool interactions, and complex agentic workflows.

• Design and enhance the agent execution loop focusing on latency, reliability, token efficiency, cost, and task completion.

• Develop and refine context management, memory/state handling, tool routing, function schemas, structured outputs, retries, and error recovery.

• Create and maintain agent evaluation infrastructure to assess quality and inform engineering decisions using data.

• Integrate and benchmark various LLM providers and models.

• Implement caching, batching, parallel tool execution, and prompt/context compression techniques.

• Establish observability and instrumentation throughout agent runs, which includes tracing, logging, metrics, and regression detection.

• Extend and adapt existing agent frameworks when current abstractions do not suffice.

• Build dependable integrations with evolving AI and tool ecosystems.

• Collaborate with product engineering teams to present clean abstractions while managing the harness complexity behind the platform.

• Troubleshoot and resolve intricate issues in non-deterministic, distributed, and model-driven systems.

⛳️ Requirements

• Over 4 years of experience in software engineering, backend engineering, or systems infrastructure.

• Strong expertise in Python and/or TypeScript.

• Practical experience in building or operating LLM-based agents in a production environment.

• Comprehensive understanding of tool calling, function schemas, context limits, structured outputs, model failures, and unpredictable LLM behavior.

• Familiarity with at least one agent framework such as LangGraph, OpenAI Agents SDK, CrewAI, AutoGen, or a custom-built agent harness.

• Robust understanding of agent orchestration and multi-step workflows.

• Experience with evaluation suites, benchmarks, A/B testing, or other measurement systems for AI products.

• Strong grasp of concurrency, caching, profiling, performance optimization, and the trade-offs between latency and cost.

• Experience with LLM APIs and production AI infrastructure.

• Exceptional debugging and problem-solving capabilities for complex and non-deterministic systems.

• A passion for technology, self-motivation, proactivity, and a strong builder mindset.

• Optional: contributions to open-source agent frameworks, LLM tools, or AI infrastructure.

• Optional: experience with RAG pipelines, vector databases, or long-term memory systems.

• Optional: familiarity with MCP or similar tool-integration standards.

• Optional: experience with LLM inference infrastructure, model routing, rate limits, fallbacks, or high-volume model APIs.

• Optional: experience with LangChain, LlamaIndex, LangGraph, DSPy, or similar AI infrastructure frameworks.

• Optional: experience with Kubernetes, Docker, cloud infrastructure, or distributed systems.

• Optional: experience in developing internal developer platforms or infrastructure utilized by multiple engineering/product teams.

• Optional: strong background in observability, distributed tracing, and production reliability.

• Optional: contributions to open-source projects or personal AI infrastructure initiatives.

🏝️ Benefits

• Competitive salary and benefits package.

• A culture that fosters ownership, experimentation, learning, and data-driven engineering.

• Direct influence over the architecture and technical roadmap.

• Collaboration with a passionate and talented team.

• Opportunities to work on real production-scale AI systems.