Senior Machine Learning Engineer Karomi Technology Private Limited

3.0 - 7.0 years

0 Lacs

chennai, tamil nadu

On-site

You will be responsible for owning the full ML stack that is capable of transforming raw dielines, PDFs, and e-commerce images into a self-learning system that can read, reason about, and design packaging artwork. This includes building data-ingestion & annotation pipelines for SVG/PDF to JSON conversion, designing and modifying model heads using technologies such as LayoutLM-v3, CLIP, GNNs, and diffusion LoRAs, training & fine-tuning on GPUs, as well as shipping inference APIs and evaluation dashboards. Your daily tasks will involve close collaboration with packaging designers and a product manager, establishing you as the technical authority on all aspects of deep learning within this domain. Your key responsibilities will be divided into three main areas: **Area Tasks:** - Data & Pre-processing (40%): Writing robust Python scripts for parsing PDF, AI, SVG files, extracting text, colour separations, images, and panel polygons. Implementing tools like Ghostscript, Tesseract, YOLO, and CLIP pipelines. Automating synthetic-copy generation for ECMA dielines and maintaining vocabulary YAMLs & JSON schemas. - Model R&D (40%): Modifying LayoutLM-v3 heads, building panel-encoder pre-train models, adding Graph-Transformer & CLIP-retrieval heads, and running experiments, hyper-param sweeps, ablations to track KPIs such as IoU, panel-F1, colour recall. - MLOps & Deployment (20%): Packaging training & inference into Docker/SageMaker or GCP Vertex jobs, maintaining CI/CD, experiment tracking, serving REST/GraphQL endpoints, and implementing an active-learning loop for designer corrections. **Must-Have Qualifications:** - 5+ years of Python experience and 3+ years of deep-learning experience with PyTorch, Hugging Face. - Hands-on experience with Transformer-based vision-language models and object-detection pipelines. - Proficiency in working with PDF/SVG tool-chains, designing custom heads/loss functions, and fine-tuning pre-trained models on limited data. - Strong knowledge of Linux, GPU, graph neural networks, and relational transformers. - Proficient in Git, code review discipline, and writing reproducible experiments. **Nice-to-Have:** - Knowledge of colour science, multimodal retrieval, diffusion fine-tuning, or packaging/CPG industry exposure. - Experience with vector search tools, AWS/GCP ML tooling, and front-end technologies like Typescript/React. You will own a tool stack including DL frameworks like PyTorch, Hugging Face Transformers, torch-geometric, parsing/CV tools, OCR/detectors, retrieval tools like CLIP/ImageBind, and MLOps tools such as Docker, GitHub Actions, W&B or MLflow. In the first 6 months, you are expected to deliver a data pipeline for converting ECMA dielines and PDFs, a panel-encoder checkpoint, an MVP copy-placement model, and a REST inference service with a designer preview UI. You will report to the Head of AI or CTO and collaborate with a front-end engineer, a product manager, and two packaging-design SMEs.,

Posted 2 days ago

Apply

Start Your Job Search Today

Browse through a variety of job opportunities tailored to your skills and preferences. Filter by location, experience, salary, and more to find your perfect fit.