Senior AI Engineer at Sage Software, Inc

Led development of a multi-agent chatbot for financial planning tasks (budget forecasting, invoice management, cash flow analysis) fully integrated with Sage Intacct. Fine-tuned GPT-4 and BERT on financial documents to improve domain relevance. Implemented Retrieval-Augmented Generation (RAG) with a knowledge base using FAISS and Pinecone; integrated LangChain and GPT-4 for up-to-date insights, reducing response time and increasing accuracy. Added ReAct-style reasoning for multi-step financial planning tasks and multi-modal GraphRAG for processing reports and charts. Maintained MCP-based state management and deployed continuous fine-tuning with human feedback and RLHF. Operated production-grade MLOps with AWS SageMaker, EKS, and Docker, achieving high uptime and scalable usage. Evaluated model performance with LangSmith and OpenAI Evals and conducted A/B tests. Also developed an AI-powered fraud detection system for Sage 50 using anomaly detection in TensorFlow and AWS Lambda.

Senior AI Engineer at Sage Software, Inc

Led development of a multi-agent chatbot integrated with Sage Intacct for financial planning tasks such as budget forecasting, invoice management, and cash-flow analysis. Pre-trained and fine-tuned GPT-4 and BERT on financial documents and client queries, improving response accuracy by 25%. Implemented Retrieval-Augmented Generation (RAG) with a Knowledge Base using FAISS and Pinecone, and integrated LangChain, Langgraph, and GPT-4 to deliver up-to-date financial insights while reducing response time by 30%. Added ReAct for multi-step cash-flow reasoning and used CoT prompting to improve structure and user satisfaction by 20%. Extended to multi-modal retrieval (GraphRAG) with CLIP for visual data in reports, achieving 35% higher decision-making accuracy. Built continuous fine-tuning and RLHF pipelines with PPO/GRPO, automated retraining with AWS SageMaker and CircleCI, reducing error rates by 20%. Deployed at scale with AWS SageMaker, AWS EKS, and Docker; monitored with Prometheus, Gra

Senior AI Engineer at Sage Software, Inc

Led development of a multi-agent chatbot for financial planning tasks integrated with Sage Intacct; pre-trained and fine-tuned LLMs (GPT-4, BERT) on financial documents; implemented Retrieval-Augmented Generation (RAG) with FAISS and Pinecone; integrated ReAct for multi-step reasoning; added multi-modal GraphRAG with CLIP; built continuous fine-tuning and RLHF pipelines; deployed with AWS SageMaker, AWS EKS, and Docker; monitored with Prometheus, Grafana, and CloudWatch; achieved 99.9% uptime and improved user retention; evaluated with LangSmith and OpenAI Evals. Also developed an AI-powered fraud detection system for Sage 50 using anomaly detection and real-time inference.

NLP Engineer at Rapid Innovation

Spearheaded an agentic finance automation system automating invoice processing across SAP and Salesforce, processing over 10,000 invoices monthly and reducing manual workload by 85% with 0.8% error rate. Fine-tuned a domain-specific LLM (GPT-3.5 Turbo) with FAISS and Elasticsearch for retrieval-augmented generation. Deployed via FastAPI on Kubernetes, boosting self-resolution rates to 75% and cutting support load by 60%. Designed an autonomous task executor using reinforcement learning (PPO) to orchestrate workflows across Slack and Salesforce, reducing handoffs by 70% and turnaround by 50%. Led an AI-driven workforce scheduling and optimization platform using reinforcement learning and time-series forecasting to improve labor utilization and reduce overtime.

NLP Engineer at Rapid Innovation

Led an agentic finance automation system automating invoice processing across SAP and Salesforce; processed over 10,000 invoices monthly, reducing manual workload by 85% and error rate to 0.8%. Fine-tuned a domain-specific LLM (GPT-3.5 Turbo) for customer support and integrated FAISS/Elasticsearch for retrieval-augmented generation; deployed via FastAPI on Kubernetes, increasing self-resolution to 75% and cutting support load by 60%, with a 40% faster resolution time. Designed an autonomous task executor using PPO-based RL in PyTorch, orchestrating Slack/Salesforce workflows, reducing manual handoffs by 70% and improving turnaround time by 50%. Led a team of five engineers to build AI-driven workforce scheduling with time-series forecasting (Prophet) and RL, delivering dynamic scheduling and reducing overtime costs by 18-25% across client deployments.

NLP Engineer at Rapid Innovation

Led an agentic finance automation system automating invoice processing across SAP and Salesforce; processed over 10,000 invoices monthly, reducing manual workload by 85% and lowering error rates to 0.8%. Fine-tuned a domain-specific LLM (GPT-3.5 Turbo) with FAISS and Elasticsearch for RAG, deployed via FastAPI on Kubernetes, increasing self-resolution rates to 75% and cutting support load by 60%. Designed an autonomous task executor using PPO-based reinforcement learning to orchestrate workflows across Slack and Salesforce, reducing manual handoffs by 70% and shortening turnaround by 50%. Built an AI-driven workforce scheduling platform using RL and Prophet, integrating with ERP/calendar systems to improve labor utilization by 25% and reduce overtime by 18%.

Junior Machine Learning Engineer at LANARS

Designed and implemented a K-means clustering model for customer segmentation. Developed an NLP-based sentiment analysis model using TF-IDF and Logistic Regression. Automated product identification for e-commerce using CNNs. Built a real-time fraud detection system with Isolation Forest and Autoencoders. Improved ML performance through hyperparameter tuning and cross-validation.

Junior Machine Learning Engineer at LANARS

Built ML solutions including a K-means clustering model for customer segmentation, NLP-based sentiment analysis using TF-IDF and Logistic Regression, and computer-vision powered product identification with CNNs for e-commerce. Implemented a real-time fraud detection system with Isolation Forest and Autoencoders, and improved model performance through hyperparameter tuning and cross-validation, achieving higher accuracy and faster inference.