I build software with solid foundations
I build software with solid foundations
DevOps & MLOps engineer who designs what he ships.
1,866 contributions on my GitHub graph 🟩, green all year, until 3 AM today, when the whole thing turned orange 🟧
When you run production systems, green turning orange is how a bad night starts 😬 So I checked the monitoring 📊, then went through the logs 🔍, looking for whatever had failed.
Nothing had. Under the graph, GitHub had written "Happy Halloween!" 🎃
Happy Halloween, everyone 👻 What's your best 3 AM false alarm? 👇
🔗 github.com/Aswin-Shine
#Halloween #DevOps #GitHub #MotionDesign
1
0
150
Create my own world
🚀 Excited to launch my latest full-stack project: The Dual-Department AI SOC & Bug Bounty Command Center!
Are you tired of standard, boring security dashboards? I built a fully automated, gamified multi-agent platform that brings AI security operations to life using a 2D Pixel-Office simulation.
The system acts like a real-life virtual cybersecurity firm with 11 specialized autonomous agents, split into two distinct departments:
🛡️ 1. Cyber Crime Branch (Forensics & OSINT):
Automated EXIF metadata & GPS extraction from media files.
Email header hop tracing & SPF/DKIM verification.
Public username footprinting & threat intel reputation lookups.
🎯 2. Bug Bounty & AppSec Lab (Recon & Vulnerabilities):
Live subdomain asset mapping & JS endpoint extraction.
High-entropy regex scanning for exposed API keys & JWT secrets.
Deterministic security misconfiguration audits (CORS reflection, Clickjacking, & Origin IP leaks).
✨ Key Highlights:
Real-time 2D Visuals: Built with Next.js and Phaser.js v3, where character sprites walk to their assigned desks, work on live terminals, and take coffee breaks during idle states (synced via WebSockets at 60fps).
Zero False Positives: Powered by a LangGraph & FastAPI backend featuring a dedicated PoC Verifier Node that runs deterministic Python checks before compiling final reports.
Live Terminal Stream: Integrated xterm.js drawer to monitor agent stdout/stderr logs in real time.
🛠️ Tech Stack: Python 3.11, FastAPI, Async Socket.io, LangGraph, Next.js, Phaser.js v3, xterm.js.
Perfect for developers, security enthusiasts, or teams looking to integrate gamified multi-agent automation into their security workflows!
📂 Interested in getting the complete source code or setting this up? Check out the link below or drop a DM to grab your license! 👇
0
67
BizPilot is an intelligent, AI-powered business workflow automation agent designed to streamline routine administrative tasks, process business data, and provide instant actionable insights through a clean, interactive dashboard.
0
38
Developed an advanced, full-stack multi-agent AI orchestration system inspired by modern office simulation frameworks, tailored specifically for cybersecurity research and automated bug hunting.
The platform features a Dual-Department architecture split into a Cyber Crime Branch (Digital Forensics & OSINT) and a Bug Bounty Lab (AppSec & Recon), managed by an autonomous Chief Orchestrator ("Michael"). Equipped with deterministic verification nodes to eliminate hallucinations and ensure 100% reliable security findings.
0
88
DeepShield is an agentic deepfake and scam defense platform designed to help users and AI agents analyze suspicious calls and documents. The project combines forensic analysis with x402 micropayments on Algorand Testnet, enabling protected security APIs to be accessed through HTTP 402 payment negotiation. The goal is to provide explainable risk analysis, evidence hashing, and verifiable on-chain transaction receipts through the LoRA Algorand Testnet Explorer.
0
76
AI-first full-stack engineer for founders who ship
AI-first full-stack engineer for founders who ship
AI Engineer | RAG, AI Agents & Next.js AI SaaS MVPs
sensei: AI code review that learns your team's style.
An open-source CLI (on PyPI) I built that learns how your team reviews code and posts inline, confidence-routed comments on GitLab merge requests: high-confidence issues get flagged, low-confidence ones stay quiet. Paired with mr-principle-review-agent (npm), which enforces team-specific coding principles on every MR. Together they save ~5 hours/week of manual review.
Stack: Python, LLM APIs, GitLab CI, agentic workflows.
Code: github.com/pulkit004/sensei
0
36
rag-foundation-stone — Enterprise RAG architecture with zero-hallucination retrieval and cloud-agnostic adapters, built in TypeScript.
0
105
WeaveAI Anchor: answers grounded in your own documents.
A document-grounded retrieval and Q&A product I built at WeaveAI. It answers only from approved documents, cites the source for every answer, and refuses when it isn't sure instead of hallucinating. Built for internal docs assistants, support knowledge copilots and product documentation search.
Stack: RAG, vector DB, retrieval evals, Next.js, LLM APIs.
Try it: weaveai.dev/products/anchor
0
33
Enterprise RAG: support resolution from days to minutes.
I architected an end-to-end RAG pipeline over a large internal document base so solutions and merchant-facing teams get contextual answers instantly. Paired with an AI insights engine that generates automated analytics summaries, change detection and risk flags, replacing manual analyst reporting cycles.
Impact: serves teams supporting 3,000+ merchants; resolution time cut from days to minutes.
Stack: RAG, LangChain, AWS Bedrock, vector DB, React, Node.js.
0
36
Full-Stack Developer building AI Agents & RAG Systems
FinSense - AI-Powered NPA Prevention Platform for Banking
AI-powered, post-disbursement loan health monitoring system built for SBI-scale banking use cases. Continuously monitors borrower financial behavior after loan disbursement, scores risk on a 0–100 scale across 4 tiers using XGBoost with SHAP-based explainability, and deploys a conversational AI agent to proactively engage at-risk borrowers 60–90 days before default — instead of only scoring creditworthiness at application time like traditional systems.
Built the FastAPI backend (loan, risk, agent, and dashboard routers), automated risk re-scoring via Celery + Redis, and real-time dashboard updates via WebSockets, with a React + Tailwind frontend for officer-facing risk visualization.
0
37
SENTINEL - Edge AI Driver Assistance System
Backend Lead for an edge AI ADAS (driver assistance) system built for Indian road conditions. Multi-stage ML pipeline covering sensor fusion, scene classification, risk scoring, GradCAM/XAI explainability, and federated learning. Built the backend and deployment pipeline, integrating ONNX and PyTorch runtimes to serve model inference in production.
0
48
Financial Decision Intelligence Platform — Multi-Agent Risk Analysis
Multi-agent platform (LangGraph) that ingests SEC 10-K filings via XBRL extraction, runs Altman Z-Score and Piotroski F-Score risk models, and generates investment committee reports. Pipeline includes RAG-based risk retrieval (FAISS + Sentence Transformers), XGBoost ML inference, and Gemini API report generation — served through a FastAPI backend with /analyze, /compare, and /report endpoints for multi-company investment ranking.
Stack: Python, LangChain, LangGraph, FAISS, XGBoost, FastAPI
0
39
RepoSphere - GitHub-Inspired Version Control Platform
Full-stack, GitHub-inspired platform with JWT authentication, repository management, issue tracking, and real-time search. Built a RESTful API with user dashboards, and reduced code redundancy by 35% through modular architecture. Deployed on AWS Amplify.
0
36
Agile Fullstack Engineer
Agile Fullstack Engineer
