Senior AI-Native Systems Software Engineer, TensorRT

Description

The TensorRT team is starting an AI-native initiative to make TensorRT the default choice for out-of-framework inference worldwide. The company is building a new framework from the ground up that uses swarms of AI agents to produce high-performance, modern C++ software at very large scale.

The role suits a systems-minded C++ engineer who wants to help grow an agentic development framework, keep up with major deep learning advances, and make model onboarding much faster for users.

The company describes itself as one of the most sought-after employers in technology.

Responsibilities

• Help design and build an architecture and codebase that can grow beyond what human engineers alone could manage, with many AI agents working in parallel to generate, test and validate production-grade software.
• Raise the output you get per unit of compute by adopting and building AI-native tools, multi-agent orchestrators and codebase harnesses, so people spend their time on the most valuable work.
• Track research and industry advances such as new attention mechanisms and KV cache strategies, then direct agent swarms to prototype them and fold them into the framework.
• Make sure the latest model families, including LLMs, diffusion, audio, vision and multi-modal models, have a smooth, high-performance route to production.
• Optimize performance across Python orchestration and C++ engine internals to deliver large latency and throughput gains for key customer workloads.

Requirements

• A BS, MS or PhD in Computer Science, Computer Engineering, AI or a related field, or equivalent experience.
• At least 4 years of relevant software development experience.
• Strong modern C++ skills, including C++11/14/17 or newer and the STL, with clean, maintainable and performant code.
• Working knowledge of modern inference frameworks and of how LLMs, diffusion and multi-modal models are architected.
• An interest in how software architecture must change to support automated, agent-driven development and codebases that keep growing.
• The ability to turn high-level customer needs into concrete technical requirements and user-focused solutions.
• A record of taking customer requests to production-quality software on tight deadlines.
• Clear communication and the ability to work across internal teams and with customers.
• Nice to have: experience with AI agent orchestrators or multi-agent coding frameworks, or with building custom agentic coding harnesses for production software.
• Nice to have: CUDA programming experience or exposure to kernel generation or autotuning work.
• Nice to have: a history of quickly turning state-of-the-art papers into working prototypes, in days rather than weeks.
• Nice to have: strong skills in software performance analysis, profiling and optimization on CPU and/or GPU, with tooling used to deliver measurable gains.

Conditions

Hybrid work arrangement (LI-Hybrid).

Base salary depends on location, experience and the pay of employees in similar roles. The range is 152,000 USD to 241,500 USD for Level 3 and 184,000 USD to 287,500 USD for Level 4.

Applications are accepted at least until October 30, 2026.

This posting is for an existing vacancy.

The company uses AI tools in parts of its recruiting process.

The company is committed to an inclusive workplace and is an equal opportunity employer that does not discriminate on protected characteristics.