Hi, I'm Arun Kashyap
Applied AI Engineer
I build intelligent AI agents|
Driven by Data, Powered by AI
Applied AI Engineer with 3+ years specializing in production-grade multi-agent systems and LLM orchestration. I've built and deployed customer-facing autonomous agents using CrewAI, LangChain, and LangGraph serving 2,000+ users on cloud platforms including GCP, AWS, and Azure.
Expert in RAG architectures, prompt engineering, and scalable microservices with FastAPI and PostgreSQL. Proven track record of reducing latency by 67% and improving system reliability through rigorous evaluation frameworks. Strong Python engineer with rapid adaptability to new languages and frameworks in agile environments.
Work Experience
AI Engineer (Agentic AI Platform)
Editlingo Solutions, LLC ā¢ Edison, NJ
- Architected multi-agent AI platform (CrewAI, LangChain, LangGraph) powering HistOracle AI with 21 autonomous agents serving 2K+ users.
- Designed hybrid RAG architecture with PostgreSQL + Pinecone, achieving 67% latency reduction (6sā2s).
- Built event-driven microservices using FastAPI, GCP Cloud Run, and async orchestration with full observability stack.
- Led cross-functional product sprints, transforming AI prototypes into production automation modules.
Graduate Student Assistant (Management of AI)
Stevens Institute of Technology ā¢ Hoboken, NJ
- Developed 11+ agentic AI systems using CrewAI + LangChain, automating 85% of academic workflows for 100+ users.
- Embedded RAG pipelines benchmarked using RAGAs and LLM-as-Judge frameworks, reducing hallucinations by 40%.
- Integrated Gemini, Claude, and OpenAI models via modular prompt optimization and contextual pipeline chaining.
- Mentored student developers on LLM orchestration and reasoning-trace analysis.
Product Analyst
BlueTree Consultancy Services ā¢ Bangalore, India
- Collaborated with enterprise clients (Autoliv, PhonePe) to engineer data automation pipelines, improving efficiency by 15%.
- Designed real-time analytics dashboards in Power BI with PostgreSQL backend orchestration.
- Conducted UAT-driven deployments with zero downtime, managing API governance and audit logging.
IoT Intern
Invicto Energies ā¢ Bangalore, India
- Developed Python pipelines for smart irrigation systems processing real-time sensor data.
- Achieved 20% increase in irrigation efficiency using Regression Analysis and predictive modeling.
Featured Projects
Campus Calm AI
3rd Place WinnerAI Student Mental Wellness Assistant providing 24/7 support. Quackathon Winner at Stevens IT (Microsoft-sponsored).
Design Thinking AI Agent Cluster
Enterprise-grade multi-agent framework orchestrating five design-thinking agents with reasoning-trace dashboards.
Anti-Money Laundering Detection
Big Data pipeline processing 31M+ financial transactions using distributed computing with AUROC 0.95 / F1 0.90.
HistOracle AI Platform
2K+ UsersMulti-agent AI platform with 21 autonomous agents, hybrid RAG architecture, and event-driven microservices.