Vamsi Mukkamala

About Me

I'm currently completing my MSc in Software Systems Engineering at UCL, where I've also been working with IBM research on applying large language models to legacy code refactoring and automated software quality analysis. Prior to this, I worked at Accenture, where I gained hands-on experience in large-scale software development and delivery and also interned at Blobcity.

As an engineer, I enjoy tackling complex problems at the intersection of systems, automation, and AI — Whether it's modernising codebases, building developer tools, or turning research into production, my goal is to create software that's scalable, reliable, and genuinely impactful.

I'm actively looking for new opportunities where I can apply my skills in software engineering, AI, and system design to impactful projects — resume. If you have a project in mind or just want to chat, feel free to reach out!

Outside of work, you'll find me watching movies, trying out new games, or grinding competitive matches in Valorant.

Skills

Languages

Python
C/C++
Java
SQL
Go
Swift

Frameworks

LangChain
LangGraph
FastAPI
Django
PyTorch
Scikit-learn
Selenium
OpenCV
SpringBoot
SwiftUI
Streamlit
Ollama
vLLM

Tools

Git & Github
AWS
JIRA
Postman
Confluence
Docker
Google Colab
Dbeaver
SQLDeveloper

Databases

PostgreSQL
Oracle
MongoDB
DynamoDB
Redis

Where I've Worked

More Details ▼

Worked in a UCL x IBM Research Collaborative program.
Designed and implemented end-to-end architecture for an AI-driven legacy code remediation system, leveraging LangChain, LangGraph to build agentic workflows.
Integrated vLLM (for industry standard inference), LangSmith instrumentation, and JaCoCo test coverage analysis, optimizing agent performance and ensuring measurable quality improvements.
Provisioned and optimized project environments with vLLM, Ollama, and reproducible virtual environments; engineered a full-repository processing pipeline enabling scalable, multi-file agentic workflows.

More Details ▼

Represented UCL at various outreach events, promoting the university's programs and initiatives to prospective students.
Collaborated with faculty and staff to enhance the student experience and provide feedback on academic programs.
Assisted in the organization of workshops and seminars, fostering a sense of community among students and faculty.

More Details ▼

Developed an automated claims workflow to accurately calculate payments owed by Cigna (health insurance) across claims received from multiple medical pharmacies.
Designed and executed test strategies, test processes, and ETL automation frameworks, improving validation efficiency and reducing manual effort by 80%. Executed and monitored AWS jobs and managed S3 storage buckets.
Applied Agile methodology using JIRA and Confluence for project planning, tracking, and documentation; increased delivery speed and team collaboration by 30%, ensuring consistent project success.
Streamlined data reporting by extracting and transforming data from Oracle DB, generating concise Excel reports and delivering key insights through PowerPoint presentations, enabling data-driven decision-making by management.

More Details ▼

Contributed to the development of around 20 machine learning and deep learning model templates with the Data Science team, reducing setup time for experiments by 40% and accelerating prototyping.
Leveraged Google Colab to optimize model training, enabling faster iteration and reproducibility across projects.
Published templates on GitHub and Blobcity Cloud, increasing open-source adoption and contributing to the company's developer community engagement.

Where I've Studied

Featured Projects

Anti-Pattern Resolutor

Local-first, agentic refactoring pipeline for Java that pairs SonarQube findings with LLM-based reasoning to detect smells/anti-patterns and propose file-scoped, behaviour-preserving edits. Changes are gated by compile + tests; public interfaces are preserved; static analysis is re-run for reporting only. Run artefacts (plans, diffs, logs) are persisted for auditability.

Project made in collaboration with UCL and IBM.

Python LangChain LangGraph Ollama vLLM SonarQube JaCoCo

Zoom Bot

Built a Python automation bot that attends online classes according to the schedule on Zoom by fetching links from our student portal and updates status on Discord server via Webhooks. Adopted by the entire 2022 CSE GITAM cohort during COVID-19

Python Selenium PyAutoGUI Discord Webhooks OpenCV SQLite

Unmasking Human Faces

Designed a deep learning model for face mask removal and facial inpainting, reconstructing occluded facial regions with realistic outputs.

Python PyTorch OpenCV Detectron2 labelflow Google Colab GANS Streamlit

Pixely

Pixely is a fun project based on image modification/editing. With pixely we can add filters, backgrounds and so on. A unique concept called image steganography can be applied to an image using pixely. Steganography a technique of hiding secret data within an ordinary, non-secret, file or message in order to avoid detection. The secret data is then extracted at its destination.

Check out the App Here.

JavaScript HTML CSS Steganography

Chameleon

Chameleon is a powerful web application designed to breathe new life into legacy codebases. By leveraging the analytical power of AI, it provides instant, in-depth analysis and modernization insights for archaic file types like COBOL and JCL and transforms them into Python code (.py files). Simply upload your files and let Chameleon reveal the underlying structure, logic, and potential for transformation.
🏆 Won Best Use of Langflow at Hacking Agents Hackathon London 2025 🏆

Langflow Prompt Engineering Python COBOL

Digit Recognizer

A Simple digit recognizing web app that classifies handwritten digits by taking the user input from the canvas and also displayes the probabilities.

Check out the App Here.

Python Streamlit PyTorch ResNet MNIST

View More Projects on GitHub More Projects on GitHub

Hello!

I'm Vamsi, a software engineer focused on designing, building & automating things.

Say Hello Back! hello@mvamsi.com