AI/ML Engineer

Description :We are looking for an AI/ML Engineer to build and scale high-performance AI systems powering millions of real-time credit advisory interactions. The role involves working on LLM integrations, AI inference optimization, async architectures, and scalable AI infrastructure to deliver low-latency, resilient user experiences.This is a high-impact opportunity to work on cutting-edge AI products and directly influence the financial journeys of millions of users across India.What Youll Own :AI Infrastructure Architecture :- Design and implement asynchronous multi-agent orchestration- Own end-to-end latency from user message to AI response- Build resilient inference pipelines that gracefully degrade under load- Implement intelligent request routing and load balancing for AI workloads- Migrate critical AI conversation flows from monolith to dedicated services- Implement WebSocket/streaming infrastructure for real-time chat- Design circuit breakers and fallback strategies for AI model failures- Build observability systems for AI performance monitoring- Optimize credit data retrieval and caching strategiesWhat we are looking for :- 3-5 years building production systems handling 10K+ concurrent users- Strong experience with async/event-driven architectures- Hands-on experience scaling ML/AI inference in production- Deep understanding of Redis, caching, and CDN strategies- Experience with message queues and real-time communication protocols- Experience integrating multiple LLM/AI models in production- Hands-on with AI model serving frameworks like TensorFlow Serving or Triton- Knowledge of inference optimization techniques such as batching, caching, and quantization- Strong understanding of conversation state management and context handling- Experience debugging production issues under high AI inference loadGrowth & Opportunity :- Direct impact on customer retention through AI performance- Exposure to cutting-edge AI infrastructure challenges- Opportunity to lead and scale the AI platform team (ref:hirist.tech)