Staff AI Engineer

ChargePoint is putting AI into the monitoring, analytics and customer support layers of its EV charging network, and this role leads that work across voice AI, computer vision and conversational AI. You would architect and build production grade AI systems that improve the monitoring platform, the support experience and intelligent automation across the charging infrastructure. The engineering detail in the posting is specific enough to judge yourself against: Python with FastAPI and Django for AI services, LLM work with LangChain, LangGraph and Amazon Bedrock or OpenAI APIs, prompt engineering and RAG architectures, and Elasticsearch including vector search and hybrid BM25 plus dense embedding retrieval. Vector databases named are Qdrant, ChromaDB and Pinecone. There is a strong production emphasis on low latency inference, caching and observability, plus distributed systems and event driven patterns with Kafka or RabbitMQ. It reports to the Director of NOC Delivery and Automation.

What the posting requires:
- 8+ years of software engineering experience, with 4+ years focused on AI and ML systems in production.
- A B.Tech, B.E. or M.S. in Computer Science, Machine Learning or a related field from what the posting describes as a top tier institution.
- Deep expertise in Python, FastAPI, Django and modern backend frameworks for AI service development.
- Hands on LLM engineering: LangChain, LangGraph, Amazon Bedrock or OpenAI APIs, prompt engineering and RAG architectures.
- Strong Elasticsearch experience including vector search, hybrid search combining BM25 with dense embeddings, and semantic retrieval.
- Proficiency with vector databases such as Qdrant, ChromaDB or Pinecone, and embedding based retrieval.
- Experience building production LLM systems with a focus on low latency inference, caching strategies and observability.
- Strong foundation in distributed systems design, microservices and event driven patterns with Kafka or RabbitMQ.
- Experience with AWS or GCP, containerisation with Docker and Kubernetes, and CI/CD pipelines.
- Strong knowledge of PostgreSQL including query optimisation and schema design, plus Redis, MongoDB and message queues.
- A track record of optimising system performance with measurable improvements in latency or cost.
- Experience with voice AI systems, telephony integrations such as Genesys or SIP, and speech processing pipelines.
- Background in computer vision, image processing or visual transformer architectures.
- Data engineering experience with Airflow, DBT and analytics platforms such as ClickHouse, Trino or Iceberg.
- Experience with massive datasets, described as 50TB and above, and building scalable data pipelines.
- Hands on experience with model fine tuning techniques such as LoRA or QLoRA.
- Familiarity with RLHF style preference tuning and model alignment.
- Experience with observability and monitoring stacks covering metrics, logging and tracing.
- Certifications in generative AI, agentic AI or a related specialisation.

What the work actually looks like:
- Lead development of AI solutions across voice AI, computer vision and conversational AI.
- Architect and build production grade AI systems that enhance the monitoring and analytics platform, improve customer support, and enable intelligent automation across the charging infrastructure.
- Work with cross functional teams to design, build and deploy AI solutions affecting EV drivers and operators worldwide.

Who you report to: the Director, NOC Delivery and Automation.

Location and working pattern: Bengaluru, India.

Honest fit guidance: this is a long requirement list even by staff standards, and the voice AI plus computer vision plus LLM combination is unusually broad. Treat the first block as the real bar and the rest as a preference list. If you have shipped production LLM retrieval systems and can talk about latency and cost numbers, apply.