Lead Software Engineer - AI Agents

Job description

**About Level AI** Level AI is on a mission to turn every customer interaction into a strategic advantage. Our AI-native platform helps enterprises transform contact centers from cost centers into engines of customer intelligence, operational efficiency, and business growth. By combining advanced AI with deep domain understanding of customer experience, Level AI empowers teams to unlock actionable insights, automate workflows, and deliver more consistent, higher-quality support across the customer journey. Headquartered in Mountain View, California, Level AI is a Series C company backed by leading investors including Battery Ventures and ENIAC. Our platform leverages Large Language Models and Custom Small Language Models (SLMs) to power AI Agents across the entire CX journey—customer-facing agents, agent-assist, and backend automation—along with deep conversation analytics for QA, coaching, and insights. With a fast-growing team in India, we are building a strong engineering presence in **Noida** to drive innovation across our platform. **About the role** As a **Lead****Engineer – AI Agents**, you will play a critical role in building the scalable backend systems that power Level AI’s next-generation AI Agents. These systems operate in real-time, high-volume enterprise environments and are central to delivering intelligent, production-grade AI experiences. This role is based in **Noida (Delhi NCR)** with a **hybrid work model (2–3 days/week in office)**. You will work at the intersection of distributed systems, cloud infrastructure, and AI-powered applications—bringing agentic AI capabilities into production at scale. ### Responsibilities - - Design and build **scalable backend systems powering AI Agents** that operate in real-time enterprise environments - Develop **agent orchestration frameworks** (multi-step reasoning, tool usage, decisioning workflows) - Build systems for **agent memory, context management, and state persistence** across interactions - Architect **low-latency inference pipelines** integrating LLMs, SLMs, and external tools/services - Implement **evaluation (evals) frameworks** to measure agent performance, accuracy, and reliability - Enable **continuous improvement loops** (feedback → retraining → deployment) for AI agents in production - Design and manage **event-driven, asynchronous workflows** for complex agent tasks - Optimize systems for **high throughput, low latency, and cost-efficient inference at scale** - Build and maintain **robust APIs and service layers** (REST / gRPC) for agent capabilities - Partner closely with **Applied AI / ML teams** to productionize models and agent behaviors - Collaborate with Product and Solutions teams to **translate real customer workflows into agentic systems** - Drive best practices in **observability, monitoring, safety, and guardrails for AI systems** - Contribute to **architecture decisions for scaling multi-tenant, enterprise-grade AI platforms** ### Requirement - - 5+ years of experience in backend engineering, distributed systems, or platform engineering - Strong experience building **high-scale, production-grade backend systems** - Experience designing systems for **real-time processing, streaming, or event-driven architectures** - Strong understanding of **API design (REST, gRPC)** and microservices architectures - Experience with **databases (SQL + NoSQL)** and data modeling for high-scale systems - Hands-on experience with **Docker, Kubernetes, and cloud platforms (AWS/GCP/Azure)** - Strong fundamentals in **system design, concurrency, and performance optimization** **Strong Plus (Agent / AI focus):** - Experience working with **LLMs, conversational AI, or AI-powered products in production** - Familiarity with **agent frameworks, tool calling, or multi-step reasoning systems** - Experience building or integrating **RAG pipelines, vector databases, or retrieval systems** - Exposure to **evaluation systems (offline/online evals, A/B testing for AI systems)** - Understanding of **prompting strategies, context windows, and model behavior optimization** Experience with **real-time decisioning systems or workflow orchestration engines**