SDE 2, Machine Learning

About the Role

We are looking for an SDE 2, Machine Learning to help build the core systems behind Glance's AI shopping agent. In this role, you will work on low-latency ML inference platforms, automated evaluation harnesses, and rapid 0→1 prototyping for new AI shopping features. You will also partner directly with applied scientists and backend engineers to take experimental ML models into reliable production environments.

What You Will Own & Build

• ML Inference Platform: Build and scale systems running recommendation inference, vector search, feature store queries, and real-time data updates.
• Evaluation Harness: Develop automated offline and online evaluation frameworks, regression tests, and latency/cost tracking tools to evaluate model releases.
• Rapid Prototyping: Rapidly prototype and test new engineering ideas for the Glance AI shopping agent, turning promising concepts into scalable platform components.
• System Performance: Optimize throughput and reduce serving latency across CPU and GPU clusters for high-volume ML workloads.

What We Are Looking For

• Engineering Foundations: Strong coding skills in Python or Java, with solid background in distributed systems.
• ML Serving & Tools: Hands-on experience with model serving tools (Triton, vLLM, TorchServe), vector databases (Qdrant, Faiss), feature stores & databases.
• Evaluation Systems: Experience building testing pipelines, regression suites, or shadow deployment setups for ML services.
• Execution & Ownership: Comfortable navigating technical ambiguity, moving quickly, and shipping code in fast-paced teams.

You might thrive in this role if you have:

• BTech or MTech in Computer Science, Machine Learning, or a related technical field.
• 3+ years of software engineering experience, with 2+ years focused on ML infrastructure, model serving, or recommendation systems.
• Hands-on experience deploying recommendation pipelines, vector search, or feature stores to production.
• Proficiency in Python, system profiling, and query performance tuning.
• Experience setting up model monitoring and maintaining high-throughput production ML systems.
• Ability to collaborate closely with Applied Science teams to turn models into production services.
• Self-directed, curious, and comfortable building solutions from 0→1.