Remote ML Engineer – Governance Platform for AI Agents

enterprisedb; United States; Remote; USD 120,000 - 160,000; Full time

Benefits offered by this job

Wellness Fridays through Dec 2026

CuraLinc health & wellness resources

Job summary

EDB, a leader in Postgres-based data and AI infrastructure, is hiring an experienced ML Engineer to own the ML infrastructure spanning serving, evaluation, red-teaming, and continuous improvement for governance models. This 100% remote role focuses on turning Postgres into a first-class platform for AI agents.

You will work on low-latency serving, model registries, and continuous retraining loops while collaborating across governance and security boundaries to maintain enterprise-grade

Qualifications

• 4-7 years of professional ML software engineering experience, with hands-on work on LLM-based or agentic systems.
• Deep production ML serving experience (low-latency inference, quantization/distillation).
• Evaluation pipeline and model registry experience.
• Red-team/adversarial testing harness experience (CI-gated regression suites).
• Drift monitoring/shadow deployment experience.
• Fluency with SQL and PostgreSQL.
• Security-first mindset in audited environments.

Responsibilities

• Serving & evaluations: low-latency serving, eval pipelines, model registry.
• Research-to-production: Red-team harness with attack orchestration, scorecards, regression suites, CI gating.
• Self-improvement loop: continuous retraining, dataset curation from audit trails, gated auto-rollout.

Skills

ML engineering

LLM/agentic systems

Low-latency serving

Model registry

Red-team CI

Drift monitoring

SQL & PostgreSQL

Security in audited environments

Job description

EDB, a leader in Postgres-based data and AI infrastructure, is hiring an experienced ML Engineer to own the ML infrastructure spanning serving, evaluation, red-teaming, and continuous improvement for governance models. This 100% remote role focuses on turning Postgres into a first-class platform for AI agents.

You will work on low-latency serving, model registries, and continuous retraining loops while collaborating across governance and security boundaries to maintain enterprise-grade