Software Engineer - ML Infrastructure

The role

This person will build and scale machine learning systems for real-time perception and inference across an edge-cloud platform. The role owns data, training, deployment, and serving infrastructure for computer vision and VLM systems.

What you will do

• Designing and implementing scalable ML training and inference pipelines for perception models and VLMs.
• Developing continuous training and evaluation systems.
• Designing large-scale multi-modal data pipelines for video and sensor data.
• Creating data pipelines for ingesting, labeling, versioning, and managing multi-modal sensor datasets.
• Implementing model monitoring, A/B testing frameworks, and performance analytics.
• Collaborating with perception researchers to transition models from research to production.
• Building tools and infrastructure for distributed training, hyperparameter optimization, and experiment tracking.

Required qualifications

• Strong software engineering fundamentals in Python.
• Working proficiency in a systems language (Rust, Go, C++).
• Working proficiency with ML frameworks (PyTorch, TensorFlow) and model optimization tooling.
• Deep experience building and operating model inference systems at scale.
• Hands-on experience with distributed compute frameworks for ML and data workloads (Ray, Spark, or equivalent).
• Strong understanding of distributed systems fundamentals: partitioning, replication, backpressure, exactly-once semantics.
• Experience with vector databases (QDrant, LanceDB, or equivalent).

Nice to have

• Familiarity with LLM/VLM serving frameworks (VLLM, SGLang, TensorRT-LLM) in production.
• Familiarity with video processing, sensor fusion, or multi-modal perception systems.

About the company

Specter builds video sensors with state-of-the-art AI agents to automate the physical world. The team is small and fast growing, with members from Anduril, Tesla, Uber, and the U.S. Special Forces.