Job Description

Results-oriented AI Engineer with 2 years of hands-on experience in designing, developing, and deploying scalable AI solutions. Expertise in the Generative AI landscape, focusing on LLM orchestration, RAG (Retrieval-Augmented Generation), and Agentic workflows. Proven track record in leveraging tools like Dify and Microsoft Copilot Studio to automate enterprise processes and enhance data-driven decision-making. Skilled in Python and data engineering to bridge the gap between raw data and production-ready AI models.

Core Technical Skills

Programming & Data: Python (Pandas, NumPy, Scikit-learn), SQL, RESTful APIs.
Generative AI & LLMs: Prompt Engineering (Chain-of-Thought, Few-shot), OpenAI API, Claude (Anthropic), Llama, Hugging Face Transformers.
AI Orchestration & Tools: Dify, Microsoft Copilot Studio, LangChain, Flowise.
Vector Databases: Pinecone, Weaviate, Milvus (for RAG implementation).
Cloud & DevOps: Azure AI Services, Docker, Git, CI/CD for AI models.

LLM Orchestration: Architected and deployed production-grade AI agents using Dify, reducing manual data entry time by 40% through automated document processing.
Enterprise Copilots: Developed custom conversational agents using Microsoft Copilot Studio, integrating internal SharePoint data and Power Automate workflows to streamline employee self-service.
Prompt Engineering: Designed and optimized complex prompt templates for various use cases, including sentiment analysis, summarization, and code generation, ensuring 95%+ accuracy in model outputs.
RAG Implementation: Built a Retrieval-Augmented Generation pipeline using Python and a Vector Database to enable AI queries over 10,000+ proprietary technical documents.
Model Training & Fine-tuning: Assisted in fine-tuning open-source LLMs for niche domain tasks, improving performance on industry-specific terminology.
Data Pipeline Development: Built automated Python scripts to clean and preprocess large datasets for model training and evaluation.
Performance Monitoring: Monitored model drift and latency, implementing caching strategies to reduce API costs by 20%.
Agentic Supply Chain Tool: Used Dify and Python to create an autonomous agent capable of analyzing supplier risk by scraping news and cross-referencing internal ERP data.
Multi-Modal AI Search: Developed a tool that allows users to query image and text data simultaneously using embedding-based search techniques.