Senior Staff Software Engineer, ML Inference

The Role

We are searching for one of the absolute best ML inference engineers in the industry—someone excited to architect and scale a cutting-edge inference system that becomes the backbone of Cognitiv's ML-driven products.

In this role, you will define what inference means to Cognitiv and lead the cross-organizational effort to bring that vision to life. You'll build performance-critical systems powering real-time decision-making for some of the world's biggest brands, while helping shape the future of AI in AdTech.

This role is foundational. It is high-impact. And it is a rare opportunity to build both the system and the team around one of the most strategic technical pillars in the company.

What You'll Do

• Build Production AdTech Systems: Design and implement reliable software and infrastructure that serves large-scale machine learning models in real-world production environments.
• Optimize for Performance at Scale: Improve throughput and latency using a mix of industry-standard frameworks and custom-built solutions tailored to Cognitiv's workloads.
• Set the Vision & Influence Execution: Define the technical direction for inference initiatives, articulate a clear vision, and influence teams across the organization to align and execute against it.
• Bridge Research to Production: Identify long-term risks and emerging technical breakthroughs, partnering closely with Research, Product, and Engineering to translate ML capabilities into business impact.
• Grow the Technical Community: Mentor engineers through code reviews, design reviews, and pair programming while elevating technical collaboration across the organization.
• Set and Automate Standards: Establish best practices for coding, testing, observability, and security — and embed them into the platform through automation.

Tech Stack

• Languages: C++17+, C#, Java
• Cloud: AWS, GCP, or Azure
• Infrastructure: Terraform, Ansible, containers
• ML: PyTorch ecosystem & model serving
• Optimization: parallelism, quantization, tiling
• Hardware Acceleration: GPU inference

Who You Are

• Strong C++ Systems Engineer: 5+ years building performance-critical software in C++17 or later, with a focus on reliability, efficiency, and production quality.
• Infrastructure-Minded Builder: Comfortable working with infrastructure-as-code (Terraform, Ansible, etc.) and thinking beyond code into deployment, reproducibility, and operational scalability.
• End-to-End Owner: You naturally take services from planning and design through implementation, delegation, testing, release, and ongoing operation — and feel accountable for outcomes, not just code.
• Clear Technical Communicator: You can articulate complex technical ideas simply, shape organization-level technical narratives, and drive alignment across Engineering, Research, and Product.

Bonus Points If You Have:

• Familiar with PyTorch or equivalent ML framework
• Experience with deep learning optimization (parallelism, quantization, tiling, etc.)
• Experience with GPU/hardware acceleration (NVIDIA TensorRT, etc.)
• Experience with ML Ops technologies (model lifecycle management, ML integrated platforms, model observability, automation, etc.)
• Familiar with containerization (Docker, Kubernetes, etc.)
• Experience with advanced ML architectures (two-tower models, teacher-student learning, etc.)
• Experience with Rust
• Experience with AI development technology (AI code review, AI code assistants, etc.)

Salary: $260,000 - $320,000 USD Base Salary + Equity