AI ML Engineer

Mitigata - Full-Stack Cyber Resilience is dedicated to empowering businesses and individuals with comprehensive digital security, compliance, and insurance solutions. Backed by over 800 clients, investors, and partners, we offer services that include cutting-edge cybersecurity solutions, tailored insurance coverage, and simplified regulatory compliance. Our advanced tools and expert consultancy protect against cyber threats, financial risks, and evolving legal challenges. We also provide customized cyber insurance for individuals, safeguarding their digital identities and personal assets. At Mitigata, we integrate cybersecurity, financial protection, and strategic advisory services to ensure holistic resilience in the digital space.

Experience: 2+ years

You'll build, deploy, and maintain ML and GenAI components that ship to production. You will work alongside senior engineers on anomaly detection pipelines and LLM-powered features, owning well-scoped modules end-to-end from data prep to model serving to monitoring.

• Build and train ML models for anomaly detection, classification, and risk scoring on cybersecurity telemetry (logs, network events, user behavior).
• Implement and tune RAG pipelines chunking strategies, embedding models, vector store indexing, retrieval evaluation.
• Develop LLM-powered features using frameworks like LangChain, LlamaIndex, or Haystack, with prompt engineering and structured output handling.
• Build data pipelines for feature engineering on streaming and batch data.
• Write production-quality Python: tests, type hints, observability, clean APIs.
• Help instrument model evaluation, drift detection, and feedback loops.
• Collaborate with security engineers and product to translate use cases into ML problems.

• 2+ years of hands-on ML/AI engineering experience (internships at strong AI teams may count).
• Strong Python with PyTorch or TensorFlow; comfort with the modern ML stack (NumPy, Pandas, scikit-learn).
• Practical experience with LLMs at least one production or substantial project using OpenAI, Anthropic, Llama, Mistral, or Gemini models.
• Hands-on with RAG, embedding models (e.g., BGE, E5, OpenAI/Voyage embeddings), and at least one vector DB (Pinecone, Qdrant, Weaviate, Milvus, or pgvector).
• Familiarity with at least one orchestration framework: LangChain, LlamaIndex, or Haystack.
• Solid grasp of classical ML for tabular/time-series data gradient boosting (XGBoost/LightGBM), clustering, autoencoders, isolation forests.
• Experience deploying models behind APIs (FastAPI, Flask) and using Docker.
• SQL and basic data engineering on cloud (AWS/GCP/Azure).
• Strong fundamentals in statistics, linear algebra, and evaluation metrics.

• Exposure to streaming systems (Kafka, Kinesis, Spark Streaming).
• Familiarity with security data formats (Syslog, CEF, OCSF, STIX/TAXII, MITRE ATT&CK).
• Fine-tuning experience with LoRA/QLoRA, PEFT, Unsloth, or Axolotl.
• Experience with experiment tracking MLflow, Weights & Biases.