Senior ML Engineer - Malware Detection

New

CloudLinuxMalware detection

Workable locations: Warsaw, Masovian Voivodeship, Poland. Madrid, Community of Madrid, Spain. Tbilisi, Tbilisi, Georgia. Sofia, Sofia City Province, Bulgaria. Yerevan, Yerevan, Armenia. Podgorica, Podgorica Municipality, Montenegro, across UTC−5 to UTC+8Full-TimeSenior

Salary not disclosed

Apply NowOpens the employer's application page

More jobs in this role

Job Details

Languages: English at upper-Intermediate (B2) level or higher
Experience: 5+ years building ML systems that ran in production, including 2+ in an adversarial domain: security, abuse, fraud, or spam.
Required Skills: PythonSQLPyTorchClickhouse

Requirements

Have 5+ years building ML systems that ran in production.
Have 2+ years in an adversarial domain such as security, abuse, fraud, or spam.
Have personally trained and fine-tuned transformer models, including data preparation, code tokenization, LoRA/QLoRA or full fine-tunes, distillation, and quantization.
Demonstrate depth in gradient boosting, linear models on sparse features, calibration, and class imbalance, including judgment about when these methods outperform an LLM.
Use leakage-free evaluation splits by time and family, precision at a fixed false-positive budget, regression sets, and account for differences between offline and production results.
Have measured inference throughput and latency and fit models to CPU or memory budgets using tools such as ONNX, llama.cpp, vLLM, int8, or int4.
Have strong Python, PyTorch, and Hugging Face skills.
Use SQL on large analytical tables, such as ClickHouse or similar.
Have familiarity with PHP or JavaScript, or be willing to learn it while working with malware analysts.
Use AI coding agents daily and be able to demonstrate how.
Have English proficiency at upper-Intermediate (B2) level or higher, with particularly strong written communication for async work.

Responsibilities

Decide which files on a server deserve closer inspection within CPU and memory budgets.
Build models to distinguish malicious code from benign code at fleet scale and a sensible cost per file.
Generate detection signatures automatically from new campaigns.
Detect malware that appears malicious only in context, including phishing pages and script injections.
Route each file to the right detection engine.
Make cleanup precise so sites stay clean afterwards.
Prove each model improves detection before it ships.
Own the path from raw samples to a production model and demonstrate its impact with measurable results.

View Full Description & ApplyYou'll be redirected to the employer's site