Urgent Predigle

7 days ago


Itanagar India Predigle Full time

Job Description About The Role We are looking for a highly skilled Software Engineer to join our team and focus on the evaluation of Generative AI systems. In this role, you will design and build robust back-end services, implement evaluation pipelines for GenAI models, and work closely with data scientists to ensure our AI systems are reliable, scalable, and production-ready. This is an exciting opportunity for an engineer who thrives at the intersection of AI, back-end engineering, and system evaluation. Key Responsibilities - Design, develop, and maintain scalable back-end services to support GenAI model evaluation and deployment. - Collaborate with data scientists and ML engineers to design evaluation frameworks, pipelines, and metrics for GenAI systems. - Implement APIs and services for automated model testing, benchmarking, and result reporting. - Apply clean coding practices, design patterns, and distributed systems knowledge to build robust evaluation workflows. - Integrate evaluation tools with FastAPI, LangChain, or agentic frameworks (where applicable). - Troubleshoot, optimize, and scale back-end systems to handle large-scale evaluations. - Document processes and ensure code quality through testing and peer reviews. Requirements - Strong programming skills in Python with practical experience in building scalable services. - Solid experience with Java (Spring Boot) for developing enterprise-grade back-end applications. - Proven expertise in back-end engineering: clean code, design patterns, and distributed systems. - Experience in building and deploying APIs, microservices, and evaluation pipelines. - Strong problem-solving skills with a structured and analytical mindset. Nice To Have - Experience with FastAPI for lightweight API development. - Exposure to LangChain, agentic workflows, or GenAI frameworks. - Familiarity with LLM evaluation methodologies (A/B testing, LLM-as-a-judge, etc.). - Knowledge of cloud environments (AWS, GCP, or Azure) and containerized deployments (Docker, Kubernetes). (ref:hirist.tech)