Senior ML/LLM Engineer

AI is only as powerful as the tools it can access. Model Context Protocol (MCP) is emerging as the standard for connecting AI to tools & systems—but enterprises can’t adopt it safely without security, observability, and control. That’s where we come in.

At Anysource, we're building the AI enablement layer for modern enterprises. Our platform enables organizations to safely connect agents to their systems through MCP, with the security, auditability, and management enterprises demand. We’re a well-funded early-stage team that built Zapier's AI platform.

We're hiring an ML/LLM Engineer (US–EU time zones) to build the machine learning foundation that powers our security platform—designing custom detectors, training classifiers, and developing embeddings that keep enterprise AI safe.

Why You'll Thrive Here

• Impact: Build the ML systems that detect and prevent AI security threats for thousands of enterprise customers.

• Excellence: Work with a small team obsessed with cutting-edge ML techniques applied to real-world security problems.

• Ownership: Drive the entire ML pipeline—from data collection and feature engineering to model deployment and continuous learning.

What You'll Do

• Build custom threat detectors. Design and train ML models to detect prompt injection, data exfiltration, jailbreaks, and novel attack patterns in real-time MCP flows.

• Develop embeddings and classifiers. Create specialized embeddings for security contexts and build rule-based classifiers that can identify malicious patterns with high precision and low latency.

• Engineer training data pipelines. Build robust data collection, labeling, and augmentation systems to continuously improve model performance with real-world threat data.

• Fine-tune and deploy LLMs. Adapt foundation models for security-specific tasks, implement efficient inference pipelines, and deploy models that can run in customer VPCs with strict latency requirements.

• Create evaluation frameworks. Build comprehensive testing suites to measure detector performance, including adversarial robustness, false positive rates, and drift detection.

• Implement federated learning systems. Design privacy-preserving learning loops that improve detectors across customers without exposing sensitive data.

What We're Looking For

• 5+ years in ML engineering with hands-on experience building production ML systems, particularly in security, fraud detection, or content moderation domains.

• Deep LLM expertise. Strong understanding of transformer architectures, fine-tuning techniques, and efficient inference methods. Experience with frameworks like Transformers, vLLM, or TensorRT.

• Security ML knowledge. Experience building detectors for adversarial inputs, anomaly detection, or threat classification. Understanding of attack vectors and defense strategies.

• Strong ML fundamentals. Proficiency in feature engineering, model evaluation, hyperparameter optimization, and MLOps practices. Experience with PyTorch, scikit-learn, and modern ML tooling.

• Data engineering skills. Ability to build robust data pipelines, handle large-scale datasets, and implement efficient training and inference workflows.

• Production deployment experience. Knowledge of containerization, model serving, monitoring, and A/B testing for ML systems in production environments.

Bonus Qualifications

• Embedding and retrieval expertise. Experience with vector databases, semantic search, and building domain-specific embeddings.

• Federated learning experience. Background in privacy-preserving ML, differential privacy, or distributed training systems.

• Security research background. Published work in adversarial ML, prompt injection detection, or AI safety research.

• Real-time ML systems. Experience building low-latency inference systems that can process high-throughput streams with sub-10ms response times.

What We Offer

We provide a competitive package designed to attract and retain top talent who can work effectively with enterprise customers.

• Competitive salary and equity — compensation that reflects your expertise and customer-facing responsibilities.

• Paid time off — 4 weeks paid vacation, paid sick leave, and paid parental leave.

• Professional development — budget for conferences, courses, and certifications in AI, enterprise software, and customer success.

• Top-tier equipment — your choice of laptop and accessories to create your ideal work environment.

• Health benefits — comprehensive health, dental, and vision coverage.

• Customer interaction opportunities — work directly with innovative companies and see the immediate impact of your work.

Diversity and inclusion are fundamental to our success.

Not quite the right fit? Reach out to [email protected] with details about your experience and interests.