SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Freshworks is building next-generation AI-driven customer and employee engagement platforms. As a Principal Engineer - Machine Learning on the AI/ML Engineering team, you will drive the architecture and technical strategy for core backend services powering the Agentic AI Platform.
Key Responsibilities:
- Lead the architectural design and execution of agent runtime orchestration services, focusing on reasoning, planning, tool invocation, and low-latency execution.
- Design and establish robust, developer-friendly APIs and SDKs that enable internal engineering teams and external ecosystem partners to build, deploy, and manage complex agent workflows.
- Build and optimize stateful dialog management, shared context systems, and memory services for multi-turn, context-aware autonomous agents.
- Establish scalable agent-to-agent (A2A) communication protocols, coordination patterns, and shared memory spaces for distributed multi-agent workflows.
- Drive the transition toward cloud-native, event-driven backend services using modern microservices, service mesh, and event-streaming architectures.
- Design and implement LLM orchestration optimizations—including semantic caching, dynamic batching, token monitoring, and intelligent model routing—to ensure high performance and cost governance.
- Direct the implementation of enterprise RAG pipelines, vector database integrations (e.g., Pinecone, Weaviate, FAISS), and deep LangSmith/OpenTelemetry tracing for end-to-end evaluation and visibility.
- Establish rigorous standards for automated testing (unit, integration, load), tenant isolation (RBAC, multi-tenancy), and provide technical mentorship to elevate engineering capabilities across the team.
This role is pivotal in establishing engineering standards, scaling backend infrastructure, and mentoring senior engineers as the company expands agentic capabilities at enterprise scale.
Note: This is a hybrid role requiring in-office presence 3 days per week (Tuesday–Thursday).
Requirements:
Must Have:
- 12–18 years of progressive software engineering and machine learning system development experience, with a track record of architecting scalable enterprise backend systems.
- Deep proficiency in Java and Python, with hands-on expertise building production systems using agentic orchestration frameworks (e.g., LangChain, LangGraph, LangSmith) and workflow engines (e.g., Temporal, Airflow).
- Advanced understanding of event-driven architectures, microservices, Kubernetes, Kafka, AWS cloud infrastructure, and mixed database paradigms (PostgreSQL, Vector DBs).
- Proven ability to navigate ambiguity, solve complex architectural bottlenecks, anticipate technical risks, and connect cross-functional dependencies to deliver high-impact engineering initiatives at scale.
- Demonstrated capability in translating complex business and AI goals into actionable long-term platform roadmaps while influencing technical decisions across cross-functional engineering partners.
- Track record of identifying innovative tools, challenging assumptions, and driving technical continuous improvement across the broader engineering organization.
Nice to Have:
- Experience with AI-driven planning systems, dialog state tracking, or reinforcement learning feedback loops in production environments.
- Hands-on experience with enterprise security frameworks, including SSO, OAuth2, RBAC, and multi-tenant data isolation.
- Active involvement in open-source AI/ML projects or contributions to major LLM/orchestration libraries.
