Senior ML Engineer – Email Security

This is a fully remote position, open to applicants in Poland, +5 more countries.

📋 Description

• Train, assess, and deploy spam-detection models: compile the corpus, select the threshold, analyze the precision/recall report, and determine whether to ship or retrain.

• Develop the production aspect of the new engine using Go: including the inference service, mail-filter integration, tokenizer, and streaming message parser.

• Enhance detection quality throughout the process, from a customer's phishing report to a rule, corpus adjustment, or retraining, balancing the cost of false positives against detection efficacy.

• Maintain the corpus tooling and training infrastructure.

• Review and authorize changes to classifiers, rules, and models.

• Address inquiries from partner engineers regarding message scoring.

• Create specifications, handbooks, and post-mortems.

• Month 1: independently deliver a rule-model release.

• Month 3: integrate and operate the first component of the new engine on the staging environment.

• Month 6: launch a model while the principal engineer is absent.

⛳️ Requirements

• A minimum of 5 years of experience in delivering detection or classification systems where you had to justify the costs associated with false positives, such as in anti-spam, anti-phishing, anti-malware, fraud, or abuse scenarios.

• Experience in taking at least one ML classifier into production, including corpus construction, threshold determination, and error analysis.

• Practical experience in transformer training using PyTorch: including tokenizers, masked-language pre-training, fine-tuning, ONNX export, and CPU inference measurement.

• Proficient in Linux, shell scripting, and Python on shared servers and GPUs.

• Regular usage of AI coding agents and tools.

• Proficient in English at an upper-intermediate (B2) level or higher, with particularly strong written communication skills for technical documentation, specifications, and code reviews.

• Strongly preferred: Knowledge of email internals such as SMTP, MIME, transfer encodings, SPF/DKIM/DMARC, and how Exim, Postfix, and Dovecot manage mail to a filter.

• Strongly preferred: Experience in developing embedded models.

• Strongly preferred: Familiarity with Go or Rust for systems programming: including parsers, streaming, memory management, and concurrency.

• Strongly preferred: Knowledge of rspamd or SpamAssassin internals: rules, composites, Lua plugins, and mass-check scoring.

• Strongly preferred: Experience with multi-million-sample corpora, labeling pipelines, and annotator tools.

• Strongly preferred: Expertise in small models for constrained hardware, including quantization, distillation, and latency budgets.

• Strongly preferred: Experience with LLM-based labelers or validators to enhance annotation scalability.

• Strongly preferred: Experience with hosting panels such as cPanel, Plesk, or DirectAdmin.

🏝️ Benefits

• Emphasis on professional growth.

• Engaging and challenging projects.

• Fully remote work with flexible hours, enabling you to manage your schedule and work from any location globally.

• Paid vacation of 24 days per year, 10 national holidays, and unlimited sick leave.

• Coverage for private medical insurance.

• Reimbursement for co-working spaces and gym/sports activities.

• Education budget available.

• Opportunity to earn a reward for the most innovative idea that the company may patent.