AK

My local time

Hi there!

I'm Abhinav, a software engineer who builds AI agents and the tools around them.

About

I'm a software engineer focused on AI systems. Most recently I was on the AI team at MRI Software, where I shipped an MCP server to production, built multi-agent pipelines with LangGraph and CrewAI, and led a frontend redesign in TypeScript and React.

I studied Computer Science at Case Western Reserve University as a Davis United World College Scholar, and did machine learning research on generating and optimizing convolutional algorithms. Outside of work I build tools for myself — like KnowledgeOS, a local-first AI knowledge base for my notes.

Experience

Jul 2025 – Jul 2026Solon, Ohio

Software Engineer II, AI Team · MRI Software

  • Architected and shipped an MCP server to production, exposing live Power BI data as LLM-accessible tools so AI agents can run filtered DAX queries across 3 datasets.
  • Built a LangGraph pipeline running 6 LLM agents to generate customer-facing insights from large-scale business datasets, reducing insight turnaround time to 4 minutes.
  • Designed a CrewAI multi-agent system that automates discrepancy detection in accounting workflows, cutting manual review time by 8 hours per month.
  • Engineered backend APIs that stream ledger and transaction data from multiple internal APIs into the CrewAI reconciliation system.
  • Led a complete frontend redesign in TypeScript/React, shipping the redesigned views to production.
  • MCP
  • LangGraph
  • CrewAI
  • Power BI
  • TypeScript
  • React

May 2024 – Dec 2024Solon, Ohio

Software Engineer Intern, AI Team · MRI Software

  • Built a lease data extraction pipeline using LangChain, Azure, and OpenAI LLMs, experimenting with search, embeddings, and relevance ranking; engineered 600+ Pydantic fields, improving extraction accuracy by 30%.
  • Designed and optimized REST APIs with concurrent processing to handle large documents, cutting extraction runtime by 70%.
  • LangChain
  • Azure
  • OpenAI
  • Pydantic
  • REST APIs

Dec 2022 – May 2024Cleveland, Ohio

Machine Learning Research Assistant · Case Western Reserve University

  • Built a framework to generate SYCL implementations of convolutional algorithms from pseudocode using the OpenAI API.
  • Contributed to a dynamic mapping framework that optimizes utilization across different layers of convolutional neural networks, extracting CNN structure details with PyTorch and ONNX.
  • Python
  • PyTorch
  • ONNX
  • SYCL
  • OpenAI API

Download full résumé (PDF)

Projects

KnowledgeOS

A local-first, AI-powered knowledge base that lets you search your personal notes and documents in natural language.

Python · FastAPI · LangGraph · ChromaDB · SSE

Source Code

Notes Repo Project

A static knowledge-management web app that turns Markdown notes into a searchable, tag-based website with automated metadata generation.

React · TypeScript · Python · GitHub Actions · Vercel

Live DemoSource Code

Web Server Analysis

Python scripts that collect, analyze, and visualize web server performance — RTT, time to first byte, and packet loss — across 35 websites.

Python · Data Analysis · Networking

Source Code

Skills

Languages
  • Python
  • TypeScript
  • JavaScript
  • Java
  • C/C++
AI / Agents
  • LangChain
  • LangGraph
  • CrewAI
  • MCP
  • OpenAI & Azure OpenAI APIs
  • RAG
  • Embeddings
Backend / Infra
  • FastAPI
  • REST APIs
  • Docker
  • Kubernetes
  • PostgreSQL
  • Okta Auth
  • Git
  • CI/CD
Frontend
  • React
  • TypeScript
  • HTML/CSS
Tools
  • Claude Code
  • GitHub Copilot
  • Azure
  • Figma