At Krisp, we’re building an AI-first Customer Experience (CX) platform to empower agents with real-time insights and automate routine support tasks. We’re looking for a Tech Lead AI Engineer who will drive the technical direction of LLM and ML infrastructure for this product. You’ll serve as the lead engineer, setting the architecture, guiding implementation, and ensuring systems are performant, scalable, and production-ready.

This is a hands-on technical leadership role: you’ll write code, design pipelines, lead critical architectural decisions, and collaborate closely with product, infrastructure, and QA teams to ship robust AI features.

• Architect, lead, and evolve LLM-based systems for Krisp’s AI CX platform, powering features like summarization, post-call analytics, conversational assistants, and AI agents.
• Lead experimentation and implementation of LLM-based and ML-based pipelines with high performance and accuracy targets.
• Design production-grade RAG pipelines to enable contextual and accurate agent-facing chat experiences.
• Build and orchestrate AI Agents capable of autonomously handling specific customer request types.
• Establish best practices for monitoring, evaluation, and iteration—tracking hallucination rate, latency, cost, and overall model performance.
• Integrate closely with product, infra, and QA teams to ensure reliability, observability, and iteration speed of AI features.
• Mentor fellow engineers, providing technical guidance and maintaining architectural clarity and standards.
• Own continuous delivery of LLM updates, including prompt regression tracking, rollback safety, and versioning.
• Stay ahead of the curve on advances in model tuning (e.g., distillation, LoRA), open-source models, and AI agent frameworks.

• Strong software engineering foundation with expert-level Python and experience in building scalable, production-grade AI services.
• Deep understanding of LLMs and prompt engineering, including hands-on experience with retrieval-augmented generation (RAG) and prompt evaluation workflows.
• Strong ML expertise, including model lifecycle, data processing, model evaluation, and serving best practices.
• Experience with vector databases and optimizing retrieval layers across structured and unstructured customer data.
• Proven ability to design cost-efficient, low-latency inference pipelines under real-time constraints.
• Solid understanding of LLM and ML evaluation, including safety checks, hallucination detection, robustness testing, and feedback loops.
• Comfortable working in cloud-native environments (GCP preferred), including experience with GPUs, autoscaling, and observability stacks.
• Strong written and verbal communication skills to drive architectural clarity and cross-team alignment.
• Passion for building user-facing AI tools that deliver measurable value in real-world scenarios like call centers and support automation.