AI Intern

About Karomi

ManageArtworks (A flagship product brought to you by Karomi, a leading Enterprise SaaS provider) enables 4000 leading Global and Indian brands. We offer everything to get artwork projects going & manage every step of the packaging and artwork process. Companies reach markets faster with our end-end packaging & artwork management system while achieving 100% compliance.

Project Duration :  2 Months

Location :  Chennai – On Site

Overview

We are enhancing an  AI-powered packaging artwork design platform  for FMCG and pharmaceutical clients that automates the creation of compliant, visually appealing packaging designs. The role focuses on building an  intelligent layout engine  that uses  Large Language Models (LLMs) , domain-specific design heuristics, layout intelligence, and visual reasoning to determine optimal placement of text, symbols, and imagery. This system aims to streamline creative workflows while ensuring regulatory compliance and delivering high-quality, automated artwork generation.

Key Deliverables

Design and implement an  AI-driven artwork layout engine  capable of interpreting textual and visual input to determine placement of design elements.

Build  LLM-based modules  that understand packaging copy content, regulatory zones, and design intent.

Develop  rule-based and heuristic models  for spacing, hierarchy, and alignment of packaging components (e.g., ingredients, logos, nutrition tables, symbols).

Integrate  vision and layout analysis models  (e.g., LayoutLM, Detectron2, or similar architectures) to map artwork panels and guide composition.

Combine  text-to-layout reasoning  (LLM) with  visual placement algorithms  for adaptive artwork generation.

Implement and integrate the layout engine into existing  Flask / FastAPI  microservices for seamless deployment in production.

Build evaluation pipelines for layout quality, visual balance, and compliance metrics.

Document model architecture, placement logic, heuristics, and integration strategy for both technical and product teams.

Desired Candidate Profile

Must-Have Skills

Strong programming skills in  Python

Knowledge of  LLMs  and  prompt engineering  for reasoning tasks

Experience in  Computer Vision  and  Natural Language Processing (NLP)

Experience with  rule-based heuristics  or constraint-based design optimization

Understanding of  packaging artwork workflows  and regulated content placement (food, pharma, or FMCG)

Proficiency in  PyTorch / TensorFlow  for custom model development

Exposure to  Flask / FastAPI  integration and API-based architectures

Experience with  vector / geometry libraries  (e.g., Shapely, OpenCV) for layout computation

Strong debugging and rapid prototyping capabilities

Good-to-Have Skills

Familiarity with  Generative AI for design automation  (text-to-image or layout synthesis)

Background in  graphic design principles , typography, and visual hierarchy

Knowledge of  SVG / PDF parsing and rendering  for artwork files

Outcome

A functional  AI-powered layout automation engine  capable of generating packaging artworks with minimal manual intervention

Integration of  LLM reasoning + heuristic rules + vision models  for intelligent design placement

Improved  design turnaround time  and  regulatory compliance consistency

Technical proof-of-concepts for  automated artwork design and layout generation

Recommendations for  scaling the system into a full design automation product