Principal Engineer, AI

A Career at HARMAN

As a technology leader that is rapidly on the move, HARMAN is filled with people who are focused on making life better. Innovation, inclusivity and teamwork are a part of our DNA. When you add that to the challenges we take on and solve together, you'll discover that at HARMAN you can grow, make a difference and be proud of the work you do every day.

About the Role

Drive hands-on delivery of AI and Generative AI solutions that streamline workflows and deliver measurable business value-measured by hours saved and the breadth of users served. You will architect, develop, and maintain production-grade systems encompassing RAG pipelines, agentic tools, model routing, vector search, evaluation, and guardrails, and observability-all tightly integrated with internal platforms and enterprise datasets.

What You Will Do

* Automate high-impact workflows for internal stakeholders, prioritizing initiatives with the greatest time savings and broadest user reach.

* Deliver production-ready copilots and customer-facing applications for knowledge search, document summarization, intelligent recommendations, conversational analytics, and end-to-end workflow automation.

* Establish operational excellence through rigorous SLAs, latency and throughput optimization, robust safety and guardrail mechanisms, transparent evaluation frameworks, and cost-efficient inference strategies.

* Architect and develop scalable, high-performance data and AI systems that support GenAI use cases including RAG, agentic workflows, and model orchestration.

* Own the complete solution lifecycle: problem definition → rapid prototyping → rigorous evaluation → production deployment → ongoing monitoring.

* Implement guardrails (content policies, safety filters), prompt and version management, latency and throughput tuning, cost controls, load balancing, and fallback or model-routing strategies.

* Design and implement RAG pipelines over heterogeneous and often messy datasets-including requirements documents, lessons learned, business rules, and unstructured content.

* Select appropriate embedding strategies, chunking approaches, vector search configurations, rerankers, and routing policies to maximize retrieval quality.

* Develop agentic workflows leveraging Lang Chain, Llama Index, MCP, and agent-to-agent (A2A) protocols; build tooling for agentic coding use cases.

* Translate subject-matter-expert knowledge into robust, maintainable prompts; evaluate trade-offs between fine-tuning and prompt engineering.

* Work hands-on with large language models, vector databases (Pinecone, FAISS), and agent memory systems.

* Containerize applications with Docker, orchestrate with Kubernetes, and automate CI/CD pipelines; manage infrastructure as code (e.g. Terraform).

* Establish observability (Datadog, Grafana, Lang Fuse), evaluation frameworks, and model/data governance and access controls appropriate for internal enterprise environments.

* Bring experience building and maintaining data lakes and warehouses (Snowflake, Delta Lake, Big Query, MS Fabric).

* Build internal copilots and customer-facing features using React, Node.js, and Python with REST or Graph

QL backends.

* Collaborate closely with requirements, testing, validation, and platform teams; thrive in a fast-paced environment with clear, proactive communication and rapid iteration.

What You Need To Be Successful

* 8+ years of experience building production software

* Programming:
Python (FastAPI, Num Py, Pandas, scikit-learn, Pydantic, Jinja2) and Node.js; strong proficiency with APIs and distributed systems.

* Model Providers:
Working familiarity with connecting to inference providers e.g. AWS Bedrock, along with OpenAI, Anthropic, Meta/Llama, and Mistral model ecosystems.

* Data & Storage: SQL and No

SQL databases (Postgre

SQL, Dynamo

DB), Elasticsearch for search and analytics, and vector databases (Pinecone, Weaviate, FAISS, Milvus, pgvector).

* Cloud &

Infrastructure: AWS (S3, EC2, Lambda, Cloud Watch, Fargate, EKS/ECS), Azure, GCP, Databricks, Docker, Kubernetes, Terraform, CI/CD, Airflow, and Kafka.

* Operational Excellence:
Load balancing, monitoring, and alerting (Datadog, Grafana, Lang…