Building systems
that grow.
I engineer agentic AI, scalable infrastructure, and intelligent products that feel alive, intuitive, and seamlessly woven into nature.
Building Systems That Feel Like The Future
Focused on building AI systems, startup products, intelligent workflows, and experimental digital environments through fast execution, continuous experimentation, and real-world iteration.
“I like building systems that survive real-world chaos.”
Experimental
Engineering.
These are not static products. They are living systems, architectural explorations, and active infrastructure built to test the limits of intelligent automation.
Active Infrastructure
AI Verify Agent
An intelligent educational verification infrastructure automating dataset validation through XGBoost-based workflows and scalable intelligence pipelines.
Experimental Engine
Jarvis
A modular agentic AI assistant exploring autonomous orchestration layers, experimental interaction systems, and automated copilots.
Startup Tooling
Pactify
An AI-powered founder workflow platform exploring safe generation systems and operational productivity infrastructure for startups.
Research &
Verification Systems
Contributing to a government-recognized cybersecurity research initiative focused on collecting, organizing, and verifying cybersecurity learning resources across multiple domains.
The initiative involves analyzing cybersecurity courses from global platforms ranging from free to paid learning ecosystems while building structured workflows for validating course information and educational datasets.
The overall system combines:
- Educational intelligence infrastructure
- AI-assisted verification workflows
- Data analysis pipelines
- Intelligent validation systems
- Scalable automation processes
Cybersecurity Educational
Intelligence Infrastructure
A large-scale cybersecurity educational verification initiative focused on validating and organizing learning resources across multiple domains including:
Leading the data analysis and verification team responsible for validating cybersecurity educational datasets and organizing structured course intelligence workflows.
To support the verification process, an ML-assisted verification system called AI-Verify Agent was developed using XGBoost-based validation workflows capable of analyzing and validating course-related information from PDFs, structured datasets, and web-based sources.
- ▹Verify cybersecurity course information
- ▹Validate educational content accuracy
- ▹Analyze structured datasets
- ▹Process PDF-based information
- ▹Support scalable verification workflows
- ▹Improve learning resource reliability
- ▹Machine learning
- ▹Automation systems
- ▹Web data extraction
- ▹Educational intelligence workflows
- ▹Scalable validation infrastructure
Execution
& Ecosystems.
Engineering intelligent systems is only half the equation. The other half is cultivating environments where ambitious people can experiment, fail, and ship.
As the Founder and President of the Entrepreneurship Cell (E-Cell), I built and scaled an operational execution team of 30 ambitious members. We transitioned the culture entirely from theoretical academics to pure 0-to-1 startup execution.
By organizing startup-focused initiatives, securing mentorship MoUs with industry leaders, and designing collaborative execution workflows, I focus on building the human infrastructure required to scale innovation.
Lab Notes
Unfinished experiments, architecture sketches, and raw thoughts on scaling intelligent systems.
Why Agentic Loops Fail in Production
Tracking context degradation over recursive LLM calls. The memory bottleneck isn't tokens, it's state-loss in nested reasoning paths...
Orchestration vs. Automation
Automation runs a script. Orchestration negotiates uncertainty. We are building too many scripts and calling them agents...
The Data Formatting Trap
70% of AI engineering is just aggressively parsing JSON out of unpredictable stochastic generators. Let's talk about defensive schema parsing...