Machine Learning Senior Software Engineer

Akamai•Cambridge, MA

1d•$121,400 - $218,600•Hybrid

About The Position

The Akamai Inference Cloud (AIC) team is part of Akamai's Cloud Technology Group. We design and operate AI platforms enabling customers to run models with unmatched performance, compliance, and economics. This team owns the end-to-end model lifecycle—from validation and security scanning through quantization, optimization, and monitoring. We ensure every model meets rigorous standards for quality, safety, and performance. As an ML Senior Software Engineer, you will build and operate systems responsible for model validation, quantization, and safety across the AIC. You'll develop pipelines that scan models for vulnerabilities, apply quantization and optimization techniques, and build guardrails that enforce safety and compliance policies. This role requires hands-on ML experience and deep understanding of modern architectures, inference optimization, and responsible AI.

Requirements

5 years of relevant experience and a Bachelor's/Master's degree in Computer Science, Machine Learning, or a related field
Demonstrate hands-on experience with modern ML frameworks such as PyTorch, TensorFlow, or JAX, including model training, fine-tuning, and inference optimization
Show proficiency in model quantization techniques (GPTQ, AWQ, GGUF) and an understanding of how model compression impacts accuracy and latency
Have experience working with large language models (LLMs), transformer architectures, and multi-modal AI systems
Demonstrate familiarity with model safety and responsible AI practices, including content filtering, red-teaming, or adversarial robustness evaluation
Show proficiency in Python and experience building production data or ML pipelines

Nice To Haves

Experience with containerized deployments, CI/CD practices, and cloud infrastructure

Responsibilities

Developing and maintaining model validation and security scanning pipelines that assess models for quality, correctness, and vulnerabilities prior to deployment
Implementing quantization, pruning, and other optimization techniques to reduce model footprint and improve inference latency across diverse hardware configurations
Building guardrail and content safety systems that enforce compliance policies and mitigate risks such as jail breaking and prompt injection
Designing model routing and prompt management infrastructure that supports intelligent request handling across model variants
Contributing to model evaluation frameworks that measure accuracy, performance, and safety metrics across the model lifecycle