Senior Software Engineer (Agentic Platform)

About the Role & Product
We are building an Agentic Platform - a platform that helps customers deploy and operate AI agents at production scale quickly and securely. The platform provides ready-to-use services/components that agent developers can quickly integrate instead of building from scratch - significantly shortening the time from idea to a complete agent. The platform also follows open protocols to easily connect with tools, data sources, and other agents.
We believe that a solid agentic platform is built on a software engineering foundation (backend & distributed systems), with AI/agent capabilities layered on top. Therefore, this role is designed with a ratio of 70% traditional software engineering/platform engineering and 30% AI, AI agents, and related protocols.
You will work directly with the engineering team to take platform services from architecture and implementation to operation - ensuring high availability, low latency, security, and scalability.
Key Responsibilities:
Platform & Backend Engineering (-70%)
Design, implement, and maintain backend services/APIs that meet high standards for availability, latency, and security.
Build platform infrastructure components to support the agent lifecycle - from where agents run, communication between agents/tools/services, state management, to security and authorization.
Build a secure sandbox runtime where AI agents can execute generated code and tools in isolated environments - using containerization/microVMs (Docker, gVisor, Firecracker), isolation through namespaces & seccomp, resource limits (CPU/memory/network), and strict egress controls to ensure untrusted agent actions never leak to the host or other tenants.
Design data models and data flows across SQL (MySQL), NoSQL (MongoDB), and vector databases; ensure consistency, throughput, and resilience.
Integrate message brokers/event-driven backbones (Kafka, RabbitMQ, AWS SQS/SNS) for asynchronous communication between internal platform services.
Identify and resolve system issues: performance bottlenecks, memory leaks, race conditions, security vulnerabilities, and resource leaks.
Design and implement observability systems (metrics, logs, distributed tracing) to monitor and debug the platform in production environments.
Research, experiment with, and evaluate new technology solutions to address technical system challenges.
2. AI & Agentic Capabilities (-30%)
Design platform services for AI agents; proactively monitor trending AI agent capabilities in the market to ensure customers always have access to the latest capabilities.
Build and integrate MCP servers as well as adapters for the A2A protocol to connect agents with tools, data sources, and other agents.
Integrate LLM/agentic frameworks (LangChain, LangGraph, CrewAI, Strands...) into the platform in a framework-agnostic manner; design abstractions so that the platform does not depend on any specific model or framework.
Design and build agent sandbox capabilities - code/tool execution environments (code interpreter, REPL, tool-call runtime) that give AI agents the ability to run code, evaluate output, and iterate safely; support multi-language execution, persistent session state, file I/O, and deterministic reproducibility so that agents can reliably "think → execute → observe → refine."
Define and enforce sandbox security boundaries - permissions per agent/per task (read-only fs, network allow-list, approval gates for sensitive actions), and deeply integrate with the platform's authorization layer so that actions within the sandbox can be audited, traced, and revoked.
Write clients (SDK, CLI, AI agent skills) to help customers integrate with and use the platform.
Apply AI tools (coding assistants, AI code review, AI-assisted testing & debugging) to daily development and operations workflows to improve team productivity.
3. Engineering Practices & Ownership (applies to both areas)
Write clean code with unit tests and integration tests; maintain quality through code reviews.
Own features/services from design to production; proactively propose architectural improvements.
Guide and review junior/mid-level engineers; contribute to building the engineering culture.
Work closely with product, infra, and stakeholders to translate requirements into clear technical solutions.
Job Requirements
I/ Must Have:
Background and Experience
Bachelor's degree or higher in Computer Science, Engineering, or a related field.
Experience building backend services/distributed systems running in production environments; for us, years of experience are only a reference - actual capability is the deciding factor.
Strong foundation in software architecture, design patterns, distributed systems, and best practices.
Strong problem-solving, logical thinking, and analytical skills; effective communication and teamwork skills.
2. Backend & Data
Proficient in at least one of: Java (Spring Boot), Go, Python.
Proficient in SQL (MySQL) and NoSQL (MongoDB); understanding of indexing, query optimization, transactions, and data modeling.
Hands-on experience with message brokers: Kafka, RabbitMQ, ActiveMQ, or AWS SQS/SNS.
Strong understanding of REST API design, authentication/authorization (OAuth, API key, service-to-service auth), and security fundamentals.
3. Infra & DevOps
Familiar with Git, Docker, CI/CD; understanding of microservices deployment.
Understanding of observability (metrics, logs, tracing) and experience working with related tools (Prometheus/Grafana, ELK/OpenSearch, OpenTelemetry, Jaeger...).
4. AI & Agentic (30%)
Understanding of core agent architecture concepts: tool use, memory/context, orchestration loop, guardrails, multi-agent coordination.
Understanding of and experience working with MCP (Model Context Protocol) - knowing how to write an MCP server is a major advantage.
Understanding of A2A (Agent-to-Agent) and other agent protocols (ADK, ReAct).
Experience using agentic frameworks (LangChain, LangGraph, CrewAI...) and/or RAG/vector databases.
Familiar with using AI tools in daily work.
II/ Nice-to-have
Experience building or contributing to a real-world agentic platform/AI platform.
Experience writing MCP servers and integrating them into production systems.
Experience implementing end-to-end monitoring/observability systems.
Experience with Kubernetes and operating container workloads at scale.
Experience with code interpreter sandboxes, identity/authorization for agents, or semantic caching.
Experience building a secure sandbox/code execution runtime for AI agents or multi-tenant workloads - hands-on experience with container isolation (Docker, gVisor, Firecracker/microVM), namespaces & seccomp, network policies, or serverless code execution platforms.
Good English listening and speaking skills - an advantage when working with international teams, partners, and technical documentation