Artificial Intelligence Engineer

Company Description Star Software provides end-to-end automated workflows for Finance operations, including Procure-to-Pay and Order-to-Cash, as well as the management of complex documents such as Certificates of Analysis and Material Test Certificates. The company's solutions are powered by Intelligent Document Processing, Artificial Intelligence, Optical Character Recognition, and Robotic Process Automation to deliver high accuracy, efficiency, and smooth integration with existing enterprise systems. Star Software focuses on continuous improvement and user-centric innovation, helping organizations streamline critical processes and reduce manual effort. Team members contribute to technology that enables clients to concentrate on strategic initiatives and drive business success, while being part of a growing Constellation of Automation.

Role Description

We are looking for an AI Engineer to build next-generation Intelligent Document Processing (IDP) and Generative AI solutions. The ideal candidate should have strong experience in Deep Learning, Computer Vision, NLP, and backend development, with the ability to train, fine-tune, evaluate, and deploy AI models as well as build production-ready AI applications.

You will work on OCR, Document AI, object detection, information extraction, Retrieval-Augmented Generation (RAG), AI agents, and enterprise AI applications.

Responsibilities

AI Model Development

· Train, fine-tune, and evaluate Deep Learning models for Document AI applications.

· Develop object detection, document layout analysis, OCR, and information extraction models.

· Build Named Entity Recognition (NER) and NLP pipelines.

· Optimize model accuracy, latency, and inference performance.

· Perform hyperparameter tuning and experiment tracking.

· Build reusable AI pipelines for document processing.

LLM & Generative AI

· Build AI chatbots and enterprise GenAI applications.

· Develop RAG pipelines using vector databases.

· Design AI agents and multi-step workflows.

· Build Text-to-SQL solutions.

· Implement prompt engineering strategies.

· Work with embeddings, semantic search, and vector retrieval.

· Design conversation memory and context management.

· Integrate OpenAI APIs and other LLM providers.

Backend Development

· Develop scalable backend services using Python and Flask.

· Design REST APIs.

· Work with PostgreSQL, Redis, and vector databases.

· Build multi-tenant AI applications.

· Optimize SQL queries and backend performance.

Deployment & Quality

· Write unit and integration tests using PyTest.

· Monitor model performance and production systems.

· Debug AI models and backend applications.

· Collaborate with product and engineering teams to deliver production-ready AI solutions.

Required Skills

Programming

· Strong Python programming

· Strong problem-solving and logical reasoning

· SQL

· Data Structures & Algorithms

Deep Learning & Machine Learning

· PyTorch

· Hugging Face Transformers

· Transfer Learning

· Fine-tuning

· Hyperparameter Tuning

· Model Evaluation

· Dataset Preparation

· Data Augmentation

· Experiment Tracking

· Model Optimization

Computer Vision

· YOLOv8

· RT-DETR

· OCR

· Document AI

· Object Detection

· Image Processing

· Layout Analysis

· Bounding Box Annotation

NLP

· Named Entity Recognition (NER)

· Information Extraction

· Text Processing

· Token Classification

· Document Understanding

· Text Classification

Generative AI

· OpenAI APIs (or equivalent LLM providers)

· LangChain

· LangGraph

· Prompt Engineering

· AI Agents

· Multi-Agent Systems

· RAG

· Embeddings

· Vector Search

· pgvector (or similar vector databases)

· Conversation Memory

· Context Management

· Function Calling / Tool Calling

· Structured Outputs

· Text-to-SQL

Backend

· Flask

· REST APIs

· PostgreSQL

· Complex SQL Query Optimization

· Redis

· Multi-tenant Architecture

· API Design

Data Processing

· Pandas

· NumPy

· JSON Processing

· Data Validation

· Data Preprocessing

MLOps & Monitoring

· Weights & Biases (W&B)

· Experiment Tracking

· Model Versioning

· Performance Monitoring

· Logging

· Benchmarking

Tools

· Git

· GitHub

· Linux

· VS Code

· Google Colab

· Roboflow

Nice to Have

· Docker

· Kubernetes

· AWS/Azure/GCP

· FastAPI

· Kafka

· Celery

· CI/CD Pipelines

· TensorRT

· ONNX Runtime

· Triton Inference Server

Preferred Qualifications

· Experience building Intelligent Document Processing (IDP) solutions.

· Experience training custom object detection and OCR models.

· Experience developing enterprise-grade Document AI pipelines.

· Experience building production LLM applications and AI agents.

· Strong understanding of model evaluation metrics, dataset creation, and production deployment.