About the team:** The Horizontal ML Technologies team creates and develops horizontal ML technologies for Avito. One of the areas is OCR: text recognition from images. The team also solves problems related to the visual quality of photographs. OCR models help solve moderation and anti-fraud tasks, as well as extract useful information from text in photographs. A separate area is the development of an OCR platform for document recognition: the team creates classic pipelines with a detector and recognizer, as well as solutions based on VLMs. In each project, the team conducts a full ML development cycle: from data collection and labeling to model training and deployment in production of high-load services. We are looking for a CV engineer with experience in solving various computer vision tasks.

Examples of future tasks: — Develop the OCR platform: improve the recognition quality of current document types and develop solutions for new ones; — Build CV and classic OCR pipelines: train a detector and recognizer, work with text type classification, image quality, pre- and post-processing; — Develop VLM solutions: fine-tune models using LoRA, work with prompting, design solution architecture, evaluate and control quality; — Collect and label data manually and automatically, generate synthetic datasets; — Optimize model inference and deploy them into production of high-load services.

We expect you to have: — 3+ years of practical experience working with ML models in the field of computer vision, OCR experience will be an advantage; — Ability to independently lead CV projects from idea to production; — Deep understanding of CV models: specifics of CNN and transformer architectures for classification, detection, and segmentation; — Understanding of how modern VLMs work and experience solving applied tasks with them; — Broad outlook in Data Science: classic ML, NLP, CV, LLM; — Proficient in Python and PyTorch, as well as familiar with classic computer vision methods in OpenCV; — Experience in optimizing and deploying models to production; — Ability to communicate with business in simple terms, explain metrics and results.

It would be great if you have: — Experience writing and reading code in Go; — Participated in ML competitions or hackathons, won prizes; — Graduated from Avito Academy, SHAD, or AI Masters.

Working with us means: — The opportunity to influence the product: your decisions will improve the experience of millions of users; — Complex technological tasks at a large scale; — Powerful hardware: access to high-performance GPU clusters for experiments; — Strong community: a team of experts with whom you can discuss ideas and approaches; — Development: budget for training, courses, conferences, and professional literature; — Health care: voluntary medical insurance with dental coverage from day one, on-site therapist and massage therapist; — Flexibility: option to work remotely or from offices in 4 Russian cities.