
ArchRivet
AI TOOLING · MCP · SYSTEM DESIGNA second look at your backend architecture.Python / FastAPI / MCPVIEW CASE STUDYA second look at a backend design. This experimental MCP tool checks the facts an assistant supplies, flags missing details, and returns findings with evidence.
Project details
The problem
A convincing architecture description can still leave retries, durability, and overload behavior unspecified.
The approach
Local rule checks turn supplied design facts into scoped findings, questions, and workload calculations. Revised facts can be reviewed again.
The result
An experimental plugin with structured evidence for the next design discussion. It does not inspect a deployed system or certify its safety.
GraphXploit
CODE INTELLIGENCE · DEVELOPER TOOLSFollow a code change through its dependencies.Tree-sitter / SQLite / GitVIEW CASE STUDYUnderstand a code change before following every reference by hand. A local workbench maps repository relationships and Git changes to potentially affected symbols.
Project details
The problem
A small change can touch code far beyond the file being edited. That context is difficult to reconstruct in an unfamiliar repository.
The approach
Static indexing, dependency graphs, and Git impact tracing bring related symbols together. Incremental SQLite indexing avoids parsing unchanged files again.
The result
A local CLI and graph workbench. The Lite edition stays local; the AI edition uses an endpoint you configure. Static relationships do not guarantee runtime behavior.
Vacation Planner
BACKEND · GO · REST APISPreferences in. An itinerary, asynchronously.Go / Gin / REST APIVIEW CASE STUDYAn API that turns season, interests, and budget into an itinerary. A background generation job lets the client submit a request, then fetch the result by ID.
Project details
The problem
LLM generation takes time. The API needs to accept a planning request without holding the client connection open for the entire response.
The approach
Gin returns a job ID, a goroutine runs the LangChainGo request, and a second endpoint exposes completion status and the itinerary.
The result
A working asynchronous API experiment. Job results are stored in memory, so they do not survive a server restart.
AI/ML Intern
TEXMiN
Foundation.
IIT (ISM) Dhanbad
Worked on AI software testing, computer vision, and real-time monitoring workflows.
- Expanded testing across computer vision, object tracking, and access control.
- Developed a monitoring prototype and a processing architecture for video analytics and event processing.
Curiosity first.
Build, test,
improve.
I study Computer Science Engineering at Chandigarh University, specializing in AI and Machine Learning. I graduate in 2027.
I like building tools that help developers understand code, and figuring out what a backend should do when a request fails.
Based in Mohali, Punjab, India.
Explore my GitHubA few good tools.
Languages
PythonGoC++SQLBackend
FastAPIGinREST APIsMCPAI and data
TensorFlowOpenCVPostgreSQLDockerMore experiments
Good company.
Good challenges.
Hack-O-Octo 2.0
Winner
IIT Ropar GenAI Hackathon
Finalist
KPIT Sparkle
Semi-Finalist
Hacktoberfest
Super Contributor
Let's talk.
Interested in AI, backend systems, or building something useful? I'd love to hear from you.
ujjansh23@gmail.comA little
off-script.
One knight, two white pieces. There’s a fork on this board. Can you find it?