AI/ML Engineer

About the role

We are seeking an AI/ML Engineer to strengthen the intelligence layer of our platform. In this hands‑on role you will design prompts, build asynchronous pipelines, train and deploy machine‑learning models, and deliver end‑to‑end production features.

Key responsibilities

• Design, build, and orchestrate AI workflows that combine live web search, LLM scoring, and structured reasoning pipelines.
• Develop and tune prompt systems for reliable, structured LLM outputs, including scoring rubrics and hallucination mitigation.
• Train, evaluate, and improve deep‑learning and gradient‑boosting models for prediction tasks using structured data.
• Deploy models as inference endpoints, manage versioning, rollback, and monitor performance (drift, accuracy decay, data quality).
• Construct async data‑collection pipelines that gather real‑time information from external sources.
• Monitor LLM behavior in production with observability tools and continuously refine prompt quality.
• Design explainable, consistent scoring systems and improve data quality through deduplication, classification, and validation logic.

Required profile

• Strong Python programming skills, especially async/await and asyncio.
• Experience with LLM APIs and prompt engineering.
• Solid understanding of ML/deep‑learning fundamentals, feature engineering, and hyper‑parameter tuning.
• Hands‑on experience with gradient‑boosting libraries such as CatBoost, XGBoost, or LightGBM.
• Familiarity with MLOps practices: model deployment, versioning, and performance monitoring.
• Proficiency in web scraping and large‑scale data extraction.
• Comfort working with relational databases and async database drivers.
• Strong debugging abilities across multi‑step asynchronous pipelines.

Required skills

• Python (async/await, asyncio)
• LLM APIs & prompt engineering
• Gradient‑boosting models (CatBoost, XGBoost, LightGBM)
• MLOps (model serving, versioning, monitoring)
• Web scraping at scale
• Relational databases & async drivers
• Deep‑learning frameworks (PyTorch, TensorFlow) – nice to have
• Retrieval‑Augmented Generation (RAG) – nice to have
• AWS services (RDS, S3, SageMaker) – nice to have
• Hyper‑parameter optimization frameworks – nice to have