Senior ML Engineer (Python, LLM)

#LLM #remote

Senior ML Engineer (Python, LLM - must)
 
Salary is up to 7500 usd gross
 
REMOTE, full-time, long termed, b2b contract, location of the candidate and legal entity, bank account of the candidate – outside of Russia, Belarus and Turkey
 
Показать контакты

 is a cybersecurity startup building a live, production AI platform that turns raw security scan data into actionable intelligence for real customers. We're looking for a hands-on AI/ML Engineer who builds and ships production services - not a researcher, and not a pure data scientist who hands models over the wall.
You'll design, build, and operate the AI/ML systems that detect, analyze, and prioritize threats. That means working across the full spectrum of modern AI tooling and, crucially, knowing which tool the problem actually needs: when a classical ML model is the right call, when a statistical or rules-based approach is enough, and when an LLM or agentic pipeline is the better fit. You should be able to reason about that trade-off and defend it, then implement whichever you chose as clean, maintainable production code.
This is an ownership role. You'll work directly with the CTO and security domain experts to translate threat-detection needs into systems that run reliably in production.

What You'll Do
Design, build, and deploy AI/ML-driven services that detect, analyze, and mitigate cybersecurity threats in production.
Analyze large volumes of security data to surface trends, patterns, and signal - and turn those findings into models and pipelines, not just reports.
Make the right architectural call for each problem: classical ML (classification, anomaly detection, prediction), statistical methods, LLMs, or agentic/orchestration pipelines. Choose deliberately and build accordingly.
Build and optimize data and inference pipelines for scalability, latency, and accuracy.
Train, fine-tune, and extend models where it's the right approach; integrate LLM APIs and prompt/agentic workflows where that's the better fit.
Establish a systematic evaluation loop - measure model and system quality offline and in production, diagnose failure patterns, and feed findings back into the system.
Collaborate with security researchers and engineering to refine detection logic and ship it into real-world applications.
Align your work with the product backlog and business strategy.

What We're Looking For
5+ years of hands-on experience as an AI/ML Engineer or Data Scientist who ships production code - ideally in cybersecurity or an adjacent data-heavy domain.
B.Sc/M.Sc in Computer Science, Engineering, Mathematics, or a related field - or equivalent demonstrated experience.
Strong proficiency in Python and solid software-engineering fundamentals: you write clean, testable, maintainable code and have built services real users depend on.
Practical command of the data science stack - PyTorch or TensorFlow, Pandas, Scikit-learn - and the judgment to know when not to reach for deep learning.
Sound applied statistics: distributions, statistical testing, evaluation metrics, and the ability to draw defensible insight from large, messy datasets.
Experience across NLP, ML, and LLMs, with a clear sense of where each belongs in a real system.
The core skill we care most about: tool selection. You can look at a problem and reason about whether it calls for a trained ML model, a statistical method, an LLM, or an agentic pipeline - and explain the trade-offs in cost, latency, accuracy, and maintainability.
Working understanding of cybersecurity concepts - threat intelligence, network security, CVEs.
Familiarity with cloud-based AI infrastructure: Amazon Bedrock (preferred), or equivalent on Azure / GCP.
B2 English