Description :

Role Overview :

You will play a critical role in bringing our AI initiatives for products to life, ensuring they are not just functional but production-grade, secure, and maintainable.

This role requires a unique blend of Machine Learning Engineering expertise, MLOps best practices, and a deep understanding of the practical challenges in deploying generative AI systems in a secure enterprise environment.

Key Responsibilities :

- Production Application Development : Lead the end-to-end lifecycle of LLM applications, transitioning functional prototypes into robust, scalable, and resilient production systems.

- LLM API Integration & Orchestration : Design and implement robust integrations with various LLM APIs (e.g., OpenAI, Anthropic, internal models), optimizing performance, cost, and reliability.

- Prompt Engineering & Optimization : Develop, test, and refine advanced prompt engineering techniques to ensure accurate, relevant, and reliable model outputs tailored to specific business use cases.

- Context Management & RAG Implementation : Implement strategies for effective context management, including Retrieval-Augmented Generation (RAG) systems, vector databases, and memory structures to enhance model relevance and accuracy.

- Output Validation & Quality Assurance : Establish rigorous validation frameworks to automatically check and verify LLM outputs against predefined constraints, minimizing hallucinations and ensuring compliance with quality standards.

- AI Security & Risk Mitigation : Implement robust security protocols to protect against adversarial attacks, specifically focusing on prompt injection, indirect prompt injection, and SQL injection vulnerabilities within the LLM application stack.

- Production Deployment & Monitoring : Utilize MLOps principles to deploy applications across cloud infrastructures (e.g., AWS, GCP, Azure), setting up comprehensive monitoring for performance metrics, latency, token usage, and drift using tools like MLflow, Weights & Biases, or Prometheus.

Required Skills and Qualifications :

Experience : 5+ years of professional experience as an ML Engineer or MLOps Engineer, with significant experience specifically focused on deploying LLM applications into production environments (beyond just demos).

Technical Proficiency :

- Strong programming skills in Python.

- Hands-on experience with ML frameworks (e.g., PyTorch, TensorFlow) and orchestration tools (e.g., Kubeflow, Airflow).

- Proficiency with cloud platforms (AWS, GCP, or Azure) and containerization technologies (Docker, Kubernetes).

- Experience with vector databases (e.g., Pinecone, Weaviate, Chroma) and RAG architecture patterns.

- Familiarity with MLOps tools for tracking, deployment, and monitoring.

- LLM Domain Knowledge : Deep understanding of current LLM capabilities, limitations, prompt engineering best practices, and emerging security vulnerabilities in generative AI.

- Problem-Solving : Strong analytical skills with a proactive approach to troubleshooting complex production issues related to model performance, latency, and system stability.

- Communication : Excellent collaboration and communication skills, capable of working effectively within cross-functional teams (Data Scientists, Software Engineers, Security Teams).