Long-form case studies, public code, the roles that produced them, and four papers from before all of this. The thread is the same: build the system, then show the evidence.
Issue to merged pull request: groomed specs, isolated worktrees, a build gate that keeps rejected work away from humans, and a harness-agnostic .agents/ toolkit with one adapter function per tool.
Agents that read the ticket and the architectural decision record before they touch code, isolated multi-repo worktrees so parallel tickets cannot collide, and a Playwright gate under a small shared AI budget across 61 production instances.
Live timing bridged from the official SignalR feed to the browser, local SQLite history for the long off-season, and a public deployment that proves the software factory works end to end.
A Raspberry Pi and a mini PC that run agents: box-box on owned hardware, scheduled openchamber workflows from a phone, and Hedwig, a resident Discord agent with no direct brokerage access.
Open source
Updated
I extended this one rather than starting it: an Orca surface alongside the existing mux panes, and routing guidance so a subagent gets a model that suits the job.
The public evidence trail for my issue-to-merged-PR agent pipeline.
I contributed upstream to the OpenCode interface I run on my homelab.
A small Go service that keeps my homelab DNS pointed at home.
My public SMART on FHIR sandbox for the healthcare integration work I know firsthand.
Health Recovery Solutions · Rimidi platform, following acquisition
Continued with the Rimidi platform after its acquisition by HRS, supporting ongoing development, helping teammates build context, and running sprint releases across 61 production instances on AWS ECS through Jenkins
Rimidi · Rimidi platform, before acquisition
Architected polymorphic evaluation framework evaluating 28 CPT codes across 6 service lines, shipping the system in 6 weeks with real-time ingestion and batch processing
Georgia Tech · CS 6440 - Introduction to Health Informatics
Mentored students on Healthcare IT technologies including HL7 FHIR, Interoperability, Clinical Decision Support (CDS), and LLMs in healthcare
Rimidi · Data Interoperability
Developed data-standardization microservice ensuring interoperability across 8 FHIR APIs (Epic, Cerner) for 20k+ patients daily
Innovaccer · Data Platform
Built real-time monitoring microservice using Flask, Snowflake, and Kafka processing 5k+ records/min
Innovaccer · Backend - License Management
Created and maintained backend for License Management and Platform Configuration services as one of three Backend engineers
Medsupervision
Created frontend views and components for the healthcare platform using ReactJS
Georgia Tech · Atlanta, GA
Specialised in Computing Systems · GPA 4.0 / 4.0
Jaypee Institute of Information Technology · Noida, India
GPA 8.2 / 10
ACM SIGCAS Conference on Computing and Sustainable Societies
When users share information about themselves on online platforms, they allow this data to be used for various purposes. In the event of users changing their minds, it becomes a strenuous task to remove the influence of this collected data from machine learning models. We look at some early models of machine unlearning strategies that have been proposed to address this challenge.
ACM SIGCAS Conference on Computing and Sustainable Societies
With the rapidly growing numbers of people who have access to smartphones, social media usage has seen a drastic increase. Such increased screen time has left many people feeling less productive. We detail the negative effects of social media addiction, look at suggested methods of combating it, and propose a method to keep up with social media happenings without feeling the constant need to check phones repeatedly.
Transactions on Emerging Telecommunications Technologies
Various distributed ledger technologies (DLTs), such as blockchain, have evolved significantly in recent years. These technologies provide a robust and effective solution for providing confidentiality, integrity, nonrepudiation, authentication, and transparency. While blockchain has many advantages, it has various limitations as well, such as complexity, low throughput, privacy, and so on. We discuss the issues that must be considered when deciding whether to use these technologies in a given case or not.
IEEE Internet of Things Magazine
Many countries across the world have not been entirely successful in their attempts to contain the spread of the COVID-19 virus. We explore the use of blockchain to help in advancing contact tracing applications, offering enhanced data security and functionality without compromising users' privacy. We explore how blockchain-based smart contracts can be used to offer incentives to users at various stages of the contact tracing process.