Yojna Setu
A multilingual government scheme discovery system. AI extracts user profile details from conversational messages, while deterministic Java rules evaluate eligibility over 82 normalized schemes.
AI Agents · AI Security · Automation · Digital Products
I build AI agents, automation systems, and developer products. From conversational workflows and backend systems to AI security and practical digital resources.
AI Agent & Security Builder · Digital Products

B.Tech in Computer Science Engineering
SRMS CET&R, Bareilly (Affiliated to AKTU)
I build software products, practical AI agents, and automation systems with a focus on AI security. I care about building practical systems rather than AI demos — separating language extraction from deterministic backend logic.
While Java and Spring Boot remain a solid engineering foundation of my work, Python is my primary language for AI agent orchestration, tool calling, and automation. Along the way, I turn repeatable engineering patterns into open developer templates and digital resources.
Engineering background in Python, Java, and Spring Boot, combining relational data modeling with modern LLM tool integrations.
Built Yojna Setu (AI + deterministic rules), RealGuard (AI automation), and QuickEats (AI-assisted product engineering & security).
Practical systems built with AI agents, deterministic rules engines, and application security.
A multilingual government scheme discovery system. AI extracts user profile details from conversational messages, while deterministic Java rules evaluate eligibility over 82 normalized schemes.
AI automation and conversational workflows for real estate. WhatsApp conversational input → Groq parameter extraction → Java business logic → RERA lookup → in-app EMI calculation → CRM persistence → broker workflow. The LLM extracts parameters while application logic handles authoritative calculations.
AI-assisted product engineering and application security system. Highlights include server-side price recalculation, ownership validation, JWT refresh-token rotation, real-time STOMP order updates, and conversational AI assistance. Verified with 71 executable @Test methods across 23 backend test classes.
Demonstrated technologies categorized by practical implementation evidence:
Practical agent workflows, tool-calling pipelines, and retrieval-grounded generation.
Defensive controls, input/output validation, and safe data boundaries for AI applications.
Core backend engineering, API contracts, and relational persistence.
Digital resources, developer templates, and practical web architectures.
End-to-end engineering demonstrated through working code.
Building practical agent workflows with defined tools, APIs, and application-side logic.
Conversational workflows, messaging integrations, and business-process automation.
Security-focused AI workflows, input validation, data boundaries, prompt-injection defenses, and secure AI integration.
Developer templates, architecture resources, reusable backend foundations, and practical digital resources.
Backend systems and interfaces required to turn an idea into a working product.
Services are for custom development where you work with me directly. For downloadable architecture blueprints and developer resources, explore the Store.
Verified coursework, cloud credentials, and engineering job simulations.
AWS Training & Certification
Foundational certification covering AWS Cloud concepts, core services (EC2, S3, RDS), security controls, and pricing models — applied directly in deploying Yojna Setu on AWS EC2.
Walmart Global Tech (Forage)
Simulated software engineering tasks including relational system design, data structure optimization, and backend debugging.
Soft Pro
Completion certificate for backend internship developing RESTful APIs with Java, Spring Boot, Hibernate ORM, and MySQL following clean MVC architecture.
Architectural blueprints, developer boilerplates, and templates.
Resources distilled from my own working systems. Strictly zero fabricated sales or fake reviews — published openly as they are finalized.
Open case study & architectural reference for decoupled WhatsApp assistants combining Groq NLP extraction with deterministic Java rules engines.
A structured backend starter kit featuring Spring Security 6, JWT rotation, database migrations, rate limiting, and clean MVC architecture.
A typographic, Swiss-inspired Next.js portfolio template built with Tailwind CSS, micro-interactions, accessible components, and zero AI fluff.
What I'm building, exploring, and shipping right now.
Building practical agent workflows where language models interpret intent while deterministic application code executes actions.
Implementing prompt-injection defenses, schema validation, data boundaries, and keeping critical calculations out of LLM prompts.
Connecting WhatsApp and webhook triggers to relational backends for real-time lead qualification and welfare scheme discovery.
Packaging verified backend foundations, architecture monographs, and starter kits into reusable templates for developers.
I believe language models must be coupled with deterministic application logic. Conversational LLMs interpret unstructured user intent, while deterministic code and relational models enforce validation, compliance, and authoritative calculations.
Direct inquiry for software engineering roles & select projects.
Have a project in mind or want to collaborate? Send me a message with details, and I'll get back to you promptly.