Senior System Software Engineer - Local AI

Overview

Lead the development of system-level software and infrastructure for deploying, benchmarking, and qualifying AI applications and models across Linux and Windows systems powered by RTX and DGX-class GPUs, as a Senior System Software Engineer on the Local AI Inference team.

What you'll do

• Partner across organizations with benchmarking teams, inference teams, and AI application developers to define requirements, design, and maintain shared infrastructure for deploying, benchmarking, and qualifying AI workloads across inference backends such as llama.cpp, Ollama, PyTorch, vLLM, and WinML.
• Build Git-based CI infrastructure for automated builds, integration testing, and regression detection across hardware platforms, operating systems, and runtime configurations.
• Develop model management services to download, version, cache, and synchronize checkpoints and dependencies for reproducible execution.
• Design databases, backend services, and data pipelines to store and query model metadata, benchmark results, system telemetry, and execution artifacts.
• Create dashboards and reports that help engineers compare inference configurations, identify regressions, and investigate performance bottlenecks.
• Develop and operate server infrastructure, including environment provisioning, workload scheduling, resource management, and failure recovery.
• Develop autonomous AI agents for infrastructure management, workload orchestration, monitoring, and diagnostics, with appropriate access controls and execution safeguards.
• Maintain Git repositories and companion websites, keeping code, documentation, model recommendations, and evaluation results current.
• Collaborate with inference developers to implement instrumentation and debugging capabilities, resolve system issues, and improve infrastructure efficiency and scalability.

What you'll need

• B.Tech or higher degree in Computer Science, Computer Engineering, Information Technology, Software Engineering, or a related field, or equivalent experience.
• 5+ years of relevant software development, with strong programming skills in Python, C/C++, C#, or another systems programming language.
• Experience developing system software, backend services, or infrastructure supporting engineering development.
• Understanding of system APIs, multithreading, process and resource management, and systems debugging.
• Hands-on with Git-based CI, databases, SQL, and data modeling.
• Strong analytical and problem-solving skills, with the ability to independently own projects and manage competing priorities.
• Excellent written and verbal communication skills for collaboration across engineering teams.

Nice to have

• Track record of integrating or debugging AI inference frameworks and building model evaluation or benchmarking platforms.
• Proficiency working with AI agents, including tool integration and evaluation of autonomous infrastructure workflows.
• Hands-on expertise in developing telemetry and observability systems using tools such as Grafana or Kibana.
• Demonstrated ability to publish technical documentation, model recommendations, or benchmark results through companion websites.

Details

• Location: Pune, India.