Full-stack delivery
I design product flows end to end, from route architecture and UI systems to backend contracts and deployment.
Full-stack / Infrastructure / AI
Phoenix, AZ · UTC−7
Hi, I'm Manan. / Software engineer
I build full-stack products, infrastructure, and applied AI that hold up in production.
Case studies
Roles
Years
Interactive terminal / click to try
/01About me
From product and backend work to infrastructure and AI, I connect the parts that make software work in production.
I'm a software engineer and graduate student at Arizona State University, studying MS in Computer Science. My background spans full-stack development, DevOps, and applied AI — built through real client work, open-source projects, and running production infrastructure on my own hardware.
I've built legal platforms, land-registry marketplaces, enterprise automation tools, self-hosted AI stacks, and developer utilities. The common thread: if something goes into production, it should be coherent to operate, observable when things go wrong, and good enough that the people using it don't notice the engineering.
“The goal isn't to use more tools. It's to remove more problems.”
I design product flows end to end, from route architecture and UI systems to backend contracts and deployment.
I build CI/CD pipelines, containerized delivery paths, and self-hosted environments that keep control close to the team.
I work with RAG, fine-tuning, local inference, and practical AI features that support real product goals.
I like tools that other developers can actually run, adapt, and trust instead of just admire in screenshots.
/02Skills
I work across product, backend, infrastructure, and AI, choosing the right tool for the job.
The stuff I reach for first when a product needs to ship, integrate, or automate.
Frontend and API work for web products with real workflows and real users.
Deployment systems that reduce manual work and make releases boring in the right way.
Applied AI work where models, retrieval, and product integration all have to cooperate.
The tools I use daily to build faster and think more clearly.
/03Featured projects
Open-source tools, client systems, and internal platforms. Each case study shows what I built, why it mattered, and how it works.
BrowserPilot
Design / code / deployment


A site-agnostic AI agent runtime that operates a real headless browser, streams a live preview to a phone, returns downloaded files, and pauses for approval before destructive actions.
Biggest challenge
The difficult boundary is not making an agent click—it is keeping identity, session state, live observation, and human approval coherent through a long-running task. The runtime must also remain testable without paying for model calls, which drove the separation between orchestration and model/browser adapters.
GitHub Orbit
Design / code / deployment
A production-ready GitHub relationship graph with progressive OAuth permissions, infinite-canvas exploration, deterministic recommendations, optional embeddings, private collections, and background sync.
Biggest challenge
Recommendations must remain useful when GitHub data is partial and rate-limited. Showing the contributing signals—and carrying coverage state through the interface—turns uncertainty into an honest product behavior instead of a hidden failure mode.




An open-source LaTeX editor for users who want live PDF preview, deployment control, and an integration-friendly API layer.
Biggest challenge
Reliable LaTeX compilation inside a self-hosted product requires handling dependency management, error propagation, and compiler state without leaning on external infrastructure. Keeping the editor state responsive while awaiting compilation output was the main UX challenge.




A polished Markdown-to-PDF workflow with live editing, themed output, and browser-based export for developers and writers.
Biggest challenge
Making PDF output visually consistent with the live preview required precise control over print-mode CSS across browsers. Achieving print-ready styling without a heavyweight editor framework was the core technical constraint.
/04Experience
I take ownership from the first technical decision through deployment and maintenance.
Aug 2025 - Present
Personal Projects
Phoenix, Arizona
Jan 2025 - Jul 2025
Balkrushna Technologies Pvt Ltd
Vadodara, India
Apr 2024 - Aug 2025
Independent Contractor
Remote
Sep 2023 - Dec 2023
Balkrushna Technologies Pvt Ltd
Vadodara, India
Nov 2021 - Jan 2022
Jalaram Wiremesh Co.
Vadodara, India
/05Education
Graduate study at ASU, building on a computer science degree from VIT and hands-on work in systems, data, and software engineering.
Arizona State University
2025 - 2027
Graduate focus on advanced systems and applied computing.
Advanced Operating Systems / Data Visualization
Vellore Institute of Technology
2021 - 2025
CGPA: 7.90 / 10.0
Artificial Intelligence / Machine Learning / Database Systems / Computer Networks / Cloud Computing
Urmi School
2019 - 2021
92.2%
Bright Day School
2018 - 2019
89.2%
/06Contact
I’m open to product, infrastructure, developer tooling, and applied AI work.
The fastest way to reach me is email. If you're working on something worth building, I'd genuinely like to hear about it.