Job Title

Role Summary You will own thefull ML stackthat turns raw dielines, PDFs, and e-commerce images into a self-learning system thatreads, reasons about, and designs packaging artwork .That includes: building data-ingestion & annotation pipelines (SVG/PDF → JSON), designing / modifying model heads on top ofLayoutLM-v3, CLIP, GNNs, diffusion LoRAs , training & fine-tuning on GPUs, shipping inference APIs and evaluation dashboards. You’ll work day-to-day with packaging designers and a product-manager; you are the technical authority on everything deep-learning for this domain.Key ResponsibilitiesData & Pre-processing (≈ 40 %) • Write robust Python scripts to parse PDF, AI, SVG; extract text, colour separations, images, panel polygons. • Implement Ghostscript, Tesseract, YOLO, CLIP pipelines. • Automate synthetic-copy generation for ECMA dielines. • Maintain vocabulary YAMLs & JSON schemas.Model R-&-D (≈ 40 %) • Modify LayoutLM-v3 heads (panel-ID, bbox-reg, colour, contrastive). • Build panel-encoder pre-train (mask-panel prediction). • Add Graph-Transformer & CLIP-retrieval heads; optional diffusion generator. • Run experiments, hyper-param sweeps, ablations; track KPIs (IoU, panel-F1, colour recall).MLOps & Deployment (≈ 20 %) • Package training & inference into Docker/SageMaker or GCP Vertex jobs. • Maintain CI/CD, experiment tracking (Weights&Biases, MLflow). • Serve REST/GraphQL endpoints that designers and the web front-end call. • Implement active-learning loop that ingests designer corrections nightly.Must-Have Qualifications 5 + yearsPython,3 + yearsdeep-learning (PyTorch, Hugging Face). Hands-on withTransformer-based vision-language models(e.g. LayoutLM, Pix2Struct) and at least oneobject-detectionpipeline (YOLOv5/8, DETR). Comfortable hackingPDF/SVGtool-chains: PyMuPDF/pdfplumber, Ghostscript, svgpathtools, OpenCV. Experience designing custom heads / loss functions and fine-tuning large pre-trained checkpoints on limited data. Solid Linux & GPU know-how; can spin up, monitor, and profile multi-GPU jobs. Familiarity withgraph neural networksor relational transformers. Clear, idiomatic Git & code-review discipline; writes reproducible experiments.Nice-to-Have Knowledge ofcolour science(Lab, ICC profiles, Pantone tables) or print production. Prior work onmultimodal retrieval(CLIP, ImageBind) ordiffusion fine-tuning(LoRA, ControlNet). Packaging / CPG industry exposure (Nutrition Facts, Drug Facts, ECMA codes). Experience standing upFAISSor similar vector search, and withAWS/GCPML tooling. Familiarity with Typescript/React front-ends for quick label-preview UIs.Primary Tool Stack You’ll Own under the following domains:DL frameworks PyTorch, Hugging Face Transformers, torch-geometricParsing / CV PyMuPDF, pdfplumber, svgpathtools, OpenCV, GhostscriptOCR / Detectors Tesseract, YOLOv8, Grounding DINO (optional)Retrieval CLIP / ImageBind + FAISSMLOps Docker, GitHub Actions, W&B or MLflow, AWS SageMaker / GCP VertexLanguages 95 % Python, occasional Bash / JSON / YAMLDeliverables in the First 6 Months Data pipeline v1that converts > 500 ECMA dielines + 200 PDFs into training-ready JSON. Panel-encoder checkpointwithMVP copy-placement model(LayoutLM-v3 backbone + heads) hitting ≥ 85 % IoU on validation. REST inference service + designer preview UI able to draft lid/side-wrap artwork for one SKU inNightly active-learning retrain loop.Reporting & Team Reports toHead of AI(or CTO). Collaborates with 1 front-end engineer, 1 product manager, 2 packaging-design SMEs.