Senior Software Engineer, Real-Time ML Serving Platform

Socket.dev; Canada; On-site; CAD 120,000 - 180,000; Full time

Job summary

Abnormal AI is seeking a Senior Software Engineer to join the Detection Team. You will architect and build scalable model serving and streaming platforms to protect against evolving AI-driven attacks, collaborating with ML engineers and data scientists.

You will own real-time pipelines, mentor engineers, and drive high-quality code reviews and design discussions, balancing rapid iteration with robust reliability and security.

Qualifications

• 5+ years of hands-on software engineering experience.
• Experience building large-scale distributed systems on cloud platforms.
• Experience with real-time data pipelines and streaming services.
• Strong cross-functional collaboration with data scientists and product teams.
• Knowledge of ML workflows and model serving at scale.
• Security and data privacy awareness.

Responsibilities

• Architect, design, build, deploy, and maintain model serving infrastructure.
• Own projects to scale model serving and data processing to 10x traffic.
• Build platform to defend against AI-based attacks.
• Own real-time and near real-time streaming pipelines and online feature serving.
• Build Abnormal’s ML Training platform to improve velocity and precision.
• Collaborate with MLE/Data Science teams to translate feedback into action.
• Mentor junior engineers through 1on1s, pair programming, and code reviews.

Skills

5+ years of experience as a Software E

Cloud platforms (AWS, GCP, Azure)

Real-time data pipelines

Cross-functional collaboration

ML workflows collaboration

Security & compliance awareness

Tools

AWS

GCP

Azure

Job description

Abnormal AI is seeking a Senior Software Engineer to join the Detection Team. You will architect and build scalable model serving and streaming platforms to protect against evolving AI-driven attacks, collaborating with ML engineers and data scientists.

You will own real-time pipelines, mentor engineers, and drive high-quality code reviews and design discussions, balancing rapid iteration with robust reliability and security.