PP/
Separate software architecture modules in a midnight workspace

The visual is unavailable. The premise still holds.

PRANJAL PANDEY / SYSTEMS ARCHITECT

Every system starts as separate parts.

The work is knowing what must connect.

One concise system. Built to hold.

Start a conversation ↗
Scroll to trace the system
01/03
00 / THE PREMISE
A WORKING PRINCIPLE

I build systems that stay useful.

Research-backed thinking, grounded in real constraints.

I work where architecture, execution, and practical AI meet. The goal is a direction your team can ship, explain, and keep improving.

CLARITY→COMPOUNDING
01 / THE WORK

Clear decisions.
Better systems.

Different products. The same question: what needs to stay true when the team, the traffic, and the requirements change?

01

System architecture

Design the shape before the scale makes it expensive.

02

Engineering execution

Turn unclear problem spaces into reliable delivery.

03

Applied AI

Move useful AI from a promising demo into a system people can trust.

Midnight software architecture control room
Work that holds up under change.
02 / HOW I WORK

Turn a hard decision
into a useful next move.

Founders and teams usually come to me when a technical choice is slowing the business down. We make the pressure visible, choose the trade-off, and leave with a path people can execute.

01 / DIAGNOSE

Find the pressure point.

Separate the real system risk from the symptom that is making the most noise.

→
02 / DECIDE

Make the trade-off explicit.

Put correctness, complexity, cost, and team capability on the same page.

→
03 / BUILD

Leave a path forward.

Turn the decision into an architecture, an artifact, or the first focused build.

Technical review table with connected architecture nodes

The output is clarity you can use.

Not a giant report. A sharper decision, a shared model of the system, and a next move the team can own.

Choose a lens to see how the decision changes.

03 / THE LONG VIEW

Keep the signal.
Lose the noise.

Pranjal Pandey in a midnight software architecture studio

Today I work where research, company building, and real systems meet.

That means staying curious about the next hard problem, while keeping the work useful to the people who have to ship it.

The long view is a practice, not a title.

2025–2028PhD in Theoretical Computer ScienceIndian Institute of Technology Jodhpur
2023–2025MTech in Artificial IntelligenceIndian Institute of Technology Jodhpur
04 / THE STORY

From first build
to the next frontier.

The tools changed, the questions got harder, and the work kept moving toward useful systems.

2013 → 2026
2026
CURRENT VENTURE

Taskflo.ai

Building an enterprise platform for teams to design, run, and improve AI workflows.

2025
RESEARCH

PhD in Theoretical Computer Science

Specialising in multi-agent systems and computational social choice at IIT Jodhpur.

2023
AI JOURNEY

MTech in Artificial Intelligence

Started a deeper study of AI systems, their limits, and how to make them work in practice.

2020
INDEPENDENT

Enterprise systems

Started working with enterprise companies as an independent contractor across architecture and execution.

2018
PRODUCT BUILDING

Mirakia

Helped build Mirakia, extending the work from product ideas into systems used by real teams.

2017
FOUNDER-BUILDER

Ordercoro

Built the architecture for a hyper-local food delivery platform with mobile apps and backend systems.

2015
FOUNDER

Corpusvision Technologies

Founded the company and worked with multiple clients across products, platforms, and delivery.

2013
FIRST BUILD

Gleron+

Built one of India's first social networking sites and learned that the best systems start with a human reason to exist.

05 / THE FIT

For founders and teams
building what comes next.

I am interested in ambitious products, difficult systems, and the people willing to think clearly about both. If you are building an AI company, backing one, or trying to make a complex platform work, there is probably a useful conversation to have.

I am not a fit for quick freelance output, speed at any cost, or buzzword-led demos.

Start the conversation ↗
06 / START HERE

Have a system
or a company to build?

Choose the channel that suits you. I am easiest to reach by email, LinkedIn, or GitHub.

Emailhello@ipranjal.com↗LinkedInConnect professionally↗GitHubSee the code and experiments↗