Staff AI Platform Engineer - Inference & Agentic Systems
Core
Build and operate multi-model inference platforms and autonomous agentic systems for internal teams and enterprise customers in fintech.
Role type
Staff AI Platform Engineer (Inference & Agentic Systems)
Builds
Paytm's AI inference platform (Pi) and the Agentic AI Platform runtime/orchestration tooling
Domain
Fintech, LLMs, Distributed Systems
Deliverable
production ML models | product features | infrastructure
Required skills
LLM-based agent architectures, distributed systems, LLM evaluation, TypeScript, Python, model serving, cloud platforms, security in agentic systems, technical leadership
Preferred skills
regulated industry experience, Model Context Protocol, model fine-tuning/quantization, workflow orchestration, multimodal/voice models
Technologies
TypeScript, Python, Bun, Temporal, Kubernetes, EC2, vLLM, TensorRT-LLM, Triton, AWS, GCP
Responsibilities
Build multi-model serving infrastructure, architect agentic runtime and orchestration, design multi-agent coordination systems, implement safety guardrails, define technical direction and patterns, mentor engineers
Seniority
Staff, hands-on IC with technical leadership