AI Engineer at Shipday (Silicon Valley) | Research Collaborator with Auburn University | Educator (2000+ students trained)

I'm passionate about bridging the gap between cutting-edge AI research and real-world applications. My work spans from building production-ready LLM systems that serve thousands of merchants to conducting empirical research on LLM reliability in software engineering.

Apr 2023 - Present

Building the future of AI-powered logistics:

• 🚀 Shipday Copilot: AI chatbot handling 30% of support tickets + on-demand reporting
• 🎯 Onboarding Assistant: RAG-powered guide reducing merchant setup time by 40%
• 📊 AI Insights Pipeline: Mining order/delivery data to surface KPI dashboards
• 👨‍💼 Driver Performance Evaluator: Lifting on-time deliveries by 18% through AI coaching
• 🌐 Review Autoresponder: Multilingual GPT agent for instant customer responses

Platform Integrations:

• Built integrations with PedidosYa, Stuart, MotoClick, Wix, EposNow
• Maintained Toast, WooCommerce, Shopify, GloriaFood integrations
• 99.9% uptime across all AI features

Contractual

• 📈 AI Chatbot: Course recommendation system improving satisfaction by 20%
• ⚡ CV Sorter: Advanced algorithms saving 40% HR screening time
• 🎨 CV Generator: ATS-optimized resume builder with AI assistance

Oct 2024 - Present

Investigating LLM API reliability through automated contract mining:

• Mining GitHub issues and Stack Overflow for LLM contract violations
• Building taxonomy of LLM-specific failure modes
• Goal: Enhance ML code reliability through automated contract generation

Research Methods:

• Web scraping with Python + MongoDB Atlas
• LLM screening with DeepSeek-R1 and GPT-4
• Statistical analysis with κ ≥ 0.80 reliability threshold

May 2022 - May 2023

"Rectangular Decomposed Graphs and Orthogonal Drawings"

• Introduced novel RDG graph class
• Developed linear-time algorithms for graph recognition and drawing
• Created Rectangular Composition Drawing methodology

• 2000+ students trained in prompt engineering and GenAI applications
• Save The Children consultant: 3-hour workshop across 6 countries
• YouTube creator: 16K+ subscribers, educational AI content

• B.Sc. Computer Science & Engineering | BUET | GPA: 3.63/4.0
• GRE: Quantitative 163 (63rd percentile), Verbal 149 (34th percentile)
• TOEFL: 105/120 (Advanced proficiency)

Specialized Skills:

• LLM APIs: OpenAI, Claude, Gemini, Llama, Mistral
• AI Frameworks: LangChain, LlamaIndex, LangGraph, CrewAI
• Vector Databases: ChromaDB, FAISS
• RAG Pipelines: Custom implementations for production systems

• AI for Software Engineering: Code generation, testing, debugging
• LLM Reliability: Contract mining, failure analysis, robustness
• Empirical Software Engineering: Developer experience, tool effectiveness
• Human-AI Collaboration: Developer productivity, AI-assisted workflows

Multi-agent AI system serving 10,000+ merchants globally

• Tech Stack: Python, FastAPI, LangChain, OpenAI, AWS
• Impact: 30% support reduction, 25% order increase

Automated system for discovering LLM API failure patterns

• Tech Stack: Python, MongoDB, GitHub API, Statistical Analysis
• Contribution: Novel taxonomy of LLM-specific contracts

Comprehensive GenAI training curriculum

• Impact: 2000+ students, 6-country reach
• Recognition: Save The Children partnership

"The future of software engineering lies not in replacing developers with AI, but in augmenting human creativity with intelligent automation. My work focuses on building that bridge."

• Publishing LLM contracts research findings
• Building next-generation AI agents for software development
• Mentoring the next wave of AI-native developers