Aryan Gupta
AI Engineer @ Perficient
- Role
- Technical Consultant at Perficient
- Location
- Boston, MA, US
- LinkedIn followers
- 500 followers
About Aryan Gupta
I specialize in building production-grade AI systems that bridge fast moving AI advancements with real-world enterprise applications.At Perficient, I’ve contributed to high-impact projects for Fortune 500 clients across the insurance and automotive industries - from developing AI-powered ingestion pipelines and Agentic AI dashboards using LangGraph, Haystack, and Azure OpenAI, to enhancing generative AI chatbots with voice mode via WebSocket for real-time, low-latency interactions.My experience spans the full AI engineering lifecycle which includes designing and architecting workflows or agentic solutions, integrating large language models (LLMs), deploying microservices with FastAPI and Docker, and implementing continuous evaluation frameworks like DeepEval to ensure accuracy, reliability, and scalability.I’m passionate about leveraging AI to solve complex problems, improve user experiences, and create systems that are not only intelligent, but also robust, maintainable, and enterprise-ready.
Experience
Technical Consultant
Jul 2025 — Present · Boston, MA, US
As part of a project for one of the largest global insurance brokers, I engineer production-grade AI workflows and agents. My work has spanned multiple domains and some are listed below.Document Ingestion & Understanding: I have implemented document ingestion pipelines using Haystack and Azure OpenAI to process complex insurance documents. These workflows handle OCR, classification, structured data extraction, and semantic enrichment, converting unstructured policies and reinsurance files into validated JSON schemas. I have also optimized retrieval pipelines to support high-precision search and downstream AI reasoning.Deep Research Agent: I contributed to the implementation of a configurable multi-agent AI system built in LangGraph, capable of performing complex, multi-step research tasks. I developed and integrated reasoning and retrieval agents that queried internal data sources, executed analytical workflows by dynamically generating Python code, and visualized results through on-the-fly graph creation. The system supported background task processing, enabling long-running research jobs with real-time status updates. I worked across both backend and frontend components, translating agentic architecture designs into services that powered an internal research and insights dashboard.Knowledge Graph & Contextual Retrieval Dashboard: Currently, I am also involved in developing a knowledge-driven dashboard that enhances AI reasoning using structured domain context. This includes modeling ontologies and instance relationships as triples, ingesting them into Neo4j, and building a knowledge-aware retrieval agent that leverages graph relationships for deeper, multi-hop insights and improved answer grounding.Across these initiatives, I also integrate LLM evaluation frameworks for continuous quality monitoring, and collaborate closely with SMEs and stakeholders to ensure scalable, enterprise-ready solutions aligned with real business needs.
Education
Boston University
Bachelor's degree, Computer Engineering
2020 — 2024
Welham Boys School
CBSE
The International School Bangalore
International Baccalaureate , Science
2018 — 2020
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.