Software Engineer- GenAI Developer

GenAI Developer

The Customer Success organization at Dassault Systèmes is dedicated to empowering our customers' success through unparalleled support and knowledge. As a Senior Customer Success GenAI Developer, you will serve as a technical anchor for our next-generation AI ecosystem. In this role, you will architect and implement advanced Enterprise RAG and GraphRAG systems, while contributing to the evolution of resilient Multi-Agent ReAct architectures.

This is a code-intensive position that demands a mastery of Python and a commitment to software craftsmanship. You will own the full engineering lifecycle—from high-throughput data ingestion via Apache NiFi and Kafka to the deployment of stateful, autonomous agents. As a key technical contributor, you will drive engineering rigor through Pydantic-based validation, rigorous code reviews, and the implementation of automated AI evaluation frameworks. You will also mentor junior developers, fostering a culture where AI is built as a robust, secure, and observable software discipline.

Role Description & Responsibilities:

• AI Engineering & Agentic Architecture

  • Architectural Mastery: Expert design of Enterprise RAG and GraphRAG (utilizing Neo4j) with a focus on hybrid search, query expansion, and cross-encoder reranking.

  • Agentic Frameworks: Proven success in building stateful, multi-agent systems using LangGraph, Haystack, or Microsoft Semantic Kernel, specifically focusing on ReAct (Reason + Act) loops and autonomous function calling.

  • Scientific Evaluation: Experience implementing automated evaluation frameworks (e.g., RAGAS, DeepEval, or G-Eval) within the CI/CD pipeline to scientifically measure faithfulness, relevancy, and retrieval precision.

  • Model Optimization: Hands-on experience with fine-tuning open-source LLMs (Llama, Mistral) and implementing model quantization and prompt caching to maximize throughput and minimize latency.
     

•  Software Craftsmanship & Pythonic Rigor

  • Advanced Python Engineering: Proficiency in idiomatic Python, including expert-level use of Asyncio for high-concurrency LLM calls and Pydantic (v2) for strict data validation and schema enforcement.

  • Production Quality: High commitment to writing maintainable, well-tested code. Experience conducting deep-dive code reviews that focus on architectural consistency and performance.

  • AI Security & Governance: Knowledge of AI security best practices, including Prompt Injection mitigation, PII masking (e.g., Microsoft Presidio), and the use of AI Gateways (e.g., LiteLLM, Portkey) for governance.
     

• Data Engineering & Observability

  •  Data Ingestion Orchestration: Proficiency in Apache NiFi for designing automated, complex data flows, alongside experience in data manipulation and cleaning at scale.

  • Stream Processing: Proficiency in integrating Kafka for real-time data ingestion and event-driven triggers for AI agent workflows.

  • MLOps & Lifecycle: Expert use of MLFlow for tracking experiments, versioning models, and managing the end-to-end model deployment lifecycle.

  • Full-Stack Observability: Knowledge of the Grafana (LGTM) stack, OpenTelemetry, and Traceloop to monitor agentic reasoning paths, token costs, and system health.
     

• Cloud Mastery & Performance (Highly Desirable)

  • Cloud-Native AI: Experience deploying AI workloads on AWS (EKS, Bedrock), Azure (AKS, OpenAI Service), or GCP (GKE, Vertex AI).

  • Cost & Latency Engineering: Proven ability to implement Semantic Caching (e.g., Redis) and intelligent model routing to optimize the cost-to-performance ratio.

  • Infrastructure as Code (IaC): Familiarity with Terraform, Pulumi, or Bicep to provision cloud-scale vector stores and GPU-accelerated compute instances.