DEVELOPER TOOLING
Aevor Platform
A developer platform designed around issue-to-pull-request workflows, with a human review gate before changes are published.
CSE SRM UNIVERSITY AP
I build backend-first systems, contribute tested changes upstream, and explore applied audio AI. My focus is software engineering roles across campus placements and off-campus opportunities—with source code and research work available to review.
02 SELECTED WORK
Five source-linked projects across developer tooling, collaboration, travel, reliability, and event applications. Open the implementation to assess architecture and code—not just a screenshot.
View all projects Scroll to exploreDEVELOPER TOOLING
A developer platform designed around issue-to-pull-request workflows, with a human review gate before changes are published.
PROJECT ARCHIVE · 04 MORE BUILDS
A GitHub-inspired collaboration platform for repositories, branches, pull requests, reviews, and merge conflicts.
A travel-stay listing platform for exploring places, browsing photos, and sharing reviews.
An early microservices scaffold organized around inventory, order, and payment service boundaries.
A full-stack commerce project with a React storefront and a Node.js, Express, and MongoDB backend.
Smaller experiments, utilities, and ongoing work live alongside these featured builds.
Semantic-Aware Zero-Trust Audio Watermarking Framework via Transformer-Based Representation and Cryptographic Binding
A reference-based speech-watermarking study that makes embedding-location selection explicit, comparing semantic contribution and spectral suitability across the full LibriSpeech test-clean corpus.
| Method | Payload success ↑ | Mean SNR (dB) ↑ |
|---|---|---|
| Joint (Semantic + Spectral) | 100% | 75.37 |
| Semantic-only | 100% | 73.30 |
| Spectral-only | 100% | 77.16 |
| Random | 99.96% | 74.25 |
Clean-reference E1 benchmark · n = 2,620. Random placement records one payload failure (99.96% success); joint roundtrip accuracy is 100% with three replicas per bit.
IEEE Signal Processing Letters · Submitted manuscript
MATLAB-based research experience focused on nano-scale sensor-network communication, including routing and packet propagation.
MATLAB · Nano-scale sensor networks · Routing · Packet propagationOPEN SOURCE CONTRIBUTION INDEX
Choose an organization or repository owner to see the public pull requests I have opened there.
Public pull requests authored by @knockknock10. The index is generated during site deployment; pull requests created after the last deployment will appear on the next build.
05 EXPERIENCE & RESEARCH
Institutional work, a summer research internship, current research questions and upstream contributions—kept distinct, with dates and claims shown only where known.
One thread connects them: understand the system, test the change, make the result reviewable.
Exploring temporal knowledge graphs, causal reasoning and GraphRAG for non-visual situational intelligence. This is an active research direction—not a claimed publication.
Source-linked work across plugin integrations, Go services, Kubernetes/network-policy testing and regression coverage. Merged and still-open work is labelled separately in the Open Source section.
Research internship during which I continued SemBind-Audio, a semantic-aware speech-watermarking study using transformer-based audio representations, STFT-based embedding placement and SHA-256 payload binding. I also worked on a MATLAB nanosensor-network communication simulator focused on routing and packet propagation.
Worked with IRHS at SRM University AP on protocols for planned initiatives. The work connected digital outreach and marketing with operational planning: structuring repeatable workflows, documenting procedures and making the steps for future activities easier to coordinate. This is described as protocol and workflow work, not as a software-engineering job ti…
06 SKILLS
These tools map to work I can point to: Go APIs and containerized services, React applications, PostgreSQL/MongoDB data layers, and audio-ML tooling used in SemBind-Audio. Rust is marked as exploration—not claimed proficiency.
const focus = {
01platform: "Go · Gin · PostgreSQL"
02services: "APIs · Auth · Data"
03workflow: "Docker · CI/CD · Review"
04goal: "Small services, clear boundaries"
}
No invented progress percentages—just the directions I’m working through now.
07 CODE & PROBLEM SOLVING
I use algorithm practice to get better at translating constraints into clear, correct implementations—then carry that discipline into the systems I build.
Discipline in practice today creates options tomorrow.
— SANJEEV KUMARPROFILE LINKS
All values come from the saved LeetCode profile.
Day-level activity is shown only when it exists in the saved data.
The profile summary is available, but the snapshot does not include a day-by-day calendar. This panel will fill in when the data source is refreshed.
Saved contest records · profile can change.
PRACTICE DESK
One verified personal profile, alongside the public practice tools and references linked here.
08 ACHIEVEMENTS & CERTIFICATIONS
Academic progress belongs here. Credentials open in a focused gallery when the issuer and evidence can be checked. No placeholder awards, invented competition placements, or inflated counters.
CREDENTIAL ARCHIVE
Open any card to view a certificate full-screen. Add certificate images or PDFs to public/certificates/; the files are discovered automatically after the next build.
09 ABOUT ME
I’m Sanjeev Kumar, a Computer Science undergraduate at SRM University AP. I’m drawn to the parts where software becomes a system: service boundaries, reliable infrastructure, data, and the trade-offs behind a feature.
Alongside building products such as Aevor, CommitHub and Wandera, I contribute to upstream repositories and explore research questions in audio watermarking and causal-temporal knowledge graphs. I try to make implementations inspectable and claims measurable.
SRM University AP · Amaravati

MY JOURNEY
I learn by taking ideas apart, implementing the hard parts, reading what failed, and improving the result. Research and open source give me a way to test those ideas against evidence and real code.
Make the implementation inspectable. Make the claim measurable.
A working principle10 CONTACT
A codebase that needs careful work, an engineering opportunity, or a research idea worth testing? Send me the context and what you hope to make better.
START A CONVERSATION
Give me enough detail to understand the idea.
Include the project or role, the problem you’re tackling, and the next step you have in mind. It helps me respond to the substance.