Sr. Machine Learning Engineer

Insilico Search Partners; Cambridge (MA); On-site; USD 150,000 - 230,000; Full time

Job summary

Insilico Search Partners seeks an experienced ML/Platform Engineer to drive productionization of experimental models, pipelines, and inference services. You will partner with researchers, own end-to-end ML delivery, and optimize systems on a shared platform built on Kubernetes and cloud infrastructure.

You will be responsible for versioned interfaces, CI/CD pipelines, and monitoring, ensuring reliable, production-grade ML capabilities for internal teams and external partners.

Qualifications

• MS degree with 5+ years in ML engineering, MLOps, or software for ML.
• Strong Python and PyTorch skills, with debugging and productionizing models.
• Ability to read research code and reason about model behavior in production.
• Experience partnering with researchers to move experiments to production.
• Hands-on with inference services, CI/CD, experiment tracking, versioning, and monitoring.
• Experience with Docker, Kubernetes, and cloud infrastructure.
• GPU-accelerated training/inference optimization; PyTorch internals.

Responsibilities

• Own the engineering lifecycle from experimental model code to production systems.
• Build and operate inference services, batch workflows, and APIs used internally and by partners/customers.
• Define versioned interfaces across models, pipelines, and downstream applications.
• Set model delivery practices on top of shared platform systems — versioning, testing, release management, monitoring, rollback.
• Build validation/regression systems keeping production aligned with research baselines.
• Profile and optimize model performance (throughput, latency, cost) and escalate platform-level issues to Platform Engineering.
• Lead cross-functional production ML initiatives from investigation through deployment and ongoing operation.

Skills

Python

PyTorch

ML engineering

MLOps

Education

Master's degree in ML/CS

Tools

Docker

Kubernetes

CI/CD

Cloud infrastructure

Job description

Our client is a venture-backed biotech company applying AI to drug discovery, using a proprietary platform to identify novel drug targets and therapeutics from complex biological data.

Own the path from experimental model code to validated, reproducible, production-grade capabilities — model code, model-specific pipelines, inference services, and validation systems used by internal teams, partners, and customers. This is a productionization role, not a research role: researchers set the science and model design, you own the engineering path to production and its reliability. You'll need enough ML depth to debug and reason about model behavior, but won't be designing architectures or running research experiments. The shared platform underneath (Kubernetes, CI/CD, infrastructure) belongs to Platform Engineering, who you'll partner with closely

Key Responsibilities

• Own the engineering lifecycle from experimental model code to tested, maintainable, performant production systems


• Partner with researchers to understand model behavior and evaluation criteria; own the engineering path to production


• Build and operate inference services, batch workflows, and APIs used internally and by external partners/customers


• Define versioned interfaces across models, pipelines, and downstream applications


• Set model delivery practices on top of shared platform systems — versioning, testing, release management, monitoring, rollback


• Build validation/regression systems keeping production aligned with research baselines


• Profile and optimize model performance (throughput, latency, cost), escalating platform-level issues to Platform Engineering


• Lead cross-functional production ML initiatives from investigation through deployment and ongoing operation

Qualifications

• M.S. with 5+ years in ML engineering, MLOps, research engineering, or software engineering for ML


• Strong Python and PyTorch (or comparable) skills — debugging, testing, profiling, productionizing models


• Enough ML depth to read research code and reason about model behavior in a production contest


• Demonstrated experience partnering with researchers/scientists to move experimental work into production


• Hands-on with inference services, CI/CD, experiment tracking, versioning, and production monitoring


• Experience with Docker, Kubernetes, and cloud infrastructure


• GPU-accelerated training/inference optimization; PyTorch internals