Hi, I'm Sujan Maharjan Software Engineer | LLM Application | Agentic Systems
I build AI systems at the intersection of software engineering and LLM research — from multi-agent pipelines and MCP integrations to applied research on language models.
About Me
A passionate software engineer with a focus on AI and Software Engineering
I started my career building full-stack web applications — frontends, backends, databases — and somewhere along the way, I got pulled toward the problem that interested me most: making AI systems work reliably in the real world.
That shift started when I integrated ChatGPT APIs into production software and began to understand what it actually takes to build with language models, not just around them. It deepened during my internship with NSF ACCESS at UIUC/NCSA, where I built an LLM-powered tool to automate reconciliation of national cyberinfrastructure records, engineered a structured output pipeline using Pydantic to eliminate hallucinations in JSON reports, and investigated MCP server architectures for research data synchronization.
Alongside that engineering work, I've been doing research. My Master's thesis at NDSU investigates two things: proving lexical anchoring as a property of large language models through novel datasets and benchmark tasks, and demonstrating that widely-cited machine unlearning methods fail to deliver on their safety claims in practice. The paper is currently under review.
My goal is AI engineering that's grounded in both - where I can build agents and pipelines that are deployable, and understand the research well enough to reason.
Experience
My professional journey
Graduate Teaching Assistant / Graduate Researcher
- Tutored and mentored undergraduate students through clarifying concepts and assisting with problem-solving.
- Graded assignments, lab works, tests, and examinations.
- Courses supported:
- CSCI 227/228 - Computing Fundamentals in Python I & II
- CSCI 336 - Theoretical Computer Science
- Conducting thesis research on Lexical Anchoring & Machine Unlearning, by creating a novel dataset and benchmark tasks.
- Implementing and fine-tuning machine learning models using PyTorch, including distributed training on HPC clusters.
NSF ACCESS STEP Intern
- Participated in the competitive NSF ACCESS STEP Program, spanning HPC training at FIU (STEP-1) and software engineering at UIUC/NCSA (STEP-2).
- As a Software Engineering Intern, I built an AI-driven tool to automate the reconciliation of national cyberinfrastructure records (CiDeR) using Large Language Models.
- Designed and deployed a robust Python web crawler to extract and process unstructured documentation from diverse HPC resource providers.
- Engineered an LLM Auditor module utilizing Pydantic schemas to generate strictly formatted, hallucination-free JSON reports detailing database discrepancies.
- Investigated internal APIs and ACCESS-MCP server architectures, identifying critical data synchronization issues and advising stakeholders on schema updates.
- Delivered a fully packaged, reproducible Python MVP application managed with modern tooling for seamless handoff to the operations team.
Associate Software Engineer
- Contributed to developing a service, with integration of ChatGPT APIs, that analyzes the given documents and interacts in natural language.
- Collaborated to develop a robust backend system with the integration of a relational database system using Django.
- Conducted Research and Development (RnD) on highlighting text sections in PDF documents by modifying Mozilla's open-source project, PDF.js.
- Experience in version control tools like Git, along with performing code review and integration, as well as manual deployment.
- Actively participated in Agile Scrum ceremonies, including sprint planning, daily stand-ups, sprint reviews, and retrospectives, contribution
- Mentored new interns on the team on frontend frameworks such as React and Next.js, as well as documented best practices within the organization.
Software Intern
- Completed a POC of developing a user-friendly, elegant, and responsive web application with a chat interface for users to upload documents and perform chat against them.
- Collaborated with the team on designing UI/UX by adopting the concept of minimalist design.
- Successfully completed the boot-camp organized for the interns on SQL, Data Warehousing, and ETL.
React Developer Intern
- Worked alongside the designers in designing a user-friendly interface for a news portal, named Samachar-pati.
- Developed and deployed the news portal website with interfaces to create, read, and manage the news and articles.
Education
Academic Background & Achievements
Master of Science in Computer Science
- Specialization in Machine Learning
Courses
- Artificial Intelligence
- Mobile Software Engineering
- Software Requirements Definition & Analysis
- Software Development Processes
- Database Systems
- Algorithm Analysis
Bachelor of Engineering in Computer Engineering
- Focus on Software Engineering
Courses
- Artificial Intelligence
- Big Data
- Data Mining
- Database Management System
- Data Structures & Algorithm
- Object Oriented Programming
- Software Engineering
- Computer Architecture
- Distributed System
Featured Projects
Some of my recent work in AI and web development
NSF-ACCESS CiDeR GenAI Tool
During my summer internship through the NSF-ACCESS STEP program at UIUC/NCSA, I developed CiDeR GenAI — an automated auditing pipeline that keeps ACCESS cyberinfrastructure resource records synchronized with Resource Provider documentation. The core problem: CiDeR database entries were manually maintained and frequently drifted from provider guides that change continuously. I integrated the CiDeR Operations API to fetch live records, built a configurable breadth-first web crawler and documentation scraper to collect provider content at scale, and deployed an LLM auditor on JetStream2 that compares each database field against scraped documentation. To eliminate hallucination risk in structured reporting, I used Pydantic schemas to constrain all LLM outputs to strict JSON — including proposed corrections, supporting evidence, and field-level modification recommendations. I also investigated the ACCESS-MCP server architecture, identified critical data synchronization gaps, and delivered findings to the operations team. The project was delivered as a fully packaged, reproducible Python MVP with documentation for seamless handoff to operations staff.
Automated Database Normalization Checker
A Python CLI tool that analyzes live PostgreSQL schemas using user-defined functional dependencies and automatically detects and decomposes relations up to Third Normal Form (3NF). Implements formal normalization algorithms inspired by database research, bridging relational theory with real-world schema design.
Medical Chatbot using Multi-Agentic RAG
An AI-powered chatbot that provides medical information and preliminary diagnoses using advanced NLP techniques and medical knowledge bases.
Let's Create Something Amazing
I'm always open to discussing new projects, creative ideas, or opportunities to be part of something meaningful.
Get In Touch