- Company Name
- Envision Technology Solutions
- Job Title
- Generative AI Engineer
- Job Description
-
**Job title**
Generative AI Engineer (Agentic AI Developer)
**Role Summary**
Design, build, and productionize Vertex AI–based Retrieval‑Augmented Generation (RAG) systems. Own the end‑to‑end pipeline from data ingestion to agent orchestration, evaluation, and deployment, integrating with vector and graph databases to deliver reliable, cost‑effective AI solutions on Google Cloud.
**Expactations**
- Produce production‑ready RAG workloads that meet performance, cost, and quality benchmarks.
- Drive continuous improvement of retrieval and agent quality through systematic evaluation and iteration.
- Maintain high engineering standards: clean code, testing, CI/CD, observability, and security.
- Work cross‑functionally with data, product, and ops teams to define metadata, access control, and deployment processes.
**Key Responsibilities**
- Build and tune RAG pipelines: chunking, embedding generation, vector indexing (Vertex Vector Search, Pinecone, Weaviate, etc.), and retrieval/reranking logic.
- Implement agentic workflows in Python (LangGraph, LangChain, LlamaIndex, Semantic Kernel, AutoGen) with tool‑calling, planning, reflection, guardrails, and structured outputs.
- Integrate agents with graph databases (Neo4j, JanusGraph, Neptune) and map graph queries (Cypher/Gremlin).
- Develop ingestion/ETL pipelines for PDFs, documents, web pages, and internal data with proper metadata and access controls.
- Conduct evaluation: retrieval metrics, answer quality, hallucination/grounding checks using tools such as RAGAS or TruLens, and iterate for quality gains.
- Deploy via Cloud Run/GKE or other GCP services; implement monitoring, observability, cost optimization, security best practices, and automated CI/CD.
**Required Skills**
- Advanced Python programming (clean architecture, async, typing, testing, packaging).
- Proven experience with Vertex AI RAG patterns, hybrid search, reranking, chunking strategies, embeddings, and prompt/schema design.
- Hands‑on with one or more agentic frameworks (LangGraph, LangChain, LlamaIndex, Semantic Kernel, AutoGen).
- Expertise in vector search concepts and at least one production vector DB (Vertex Vector Search, Pinecone, Weaviate, Milvus, pgvector).
- Solid knowledge of graph databases, data modeling, and querying (Cypher, Gremlin, SPARQL basics).
- Familiarity with GCP fundamentals: IAM, Cloud Logging/Monitoring, Cloud Run/GKE, Cloud Storage, Pub/Sub/Kafka.
- Strong engineering practices: code reviews, unit/integration testing, telemetry, secure‑by‑design, reliability mindset.
**Required Education & Certifications**
- Bachelor’s degree in Computer Science, Software Engineering, or a related technical field.
- Optional: Google Cloud Certifications (e.g., Professional Data Engineer, Professional Cloud Developer) or equivalent cloud‑native credentials.
Berkeley heights, United states
On site
18-02-2026