Let's talk

AI Architect & Founder

Divya Prakash Jha

An agentic AI architect building decision intelligence and automated business assistants from Palghar, Maharashtra.

By Divyaprakash JhaUpdated 5 min readTopic: Founder

Photo for Divya Prakash Jha

Divya Prakash Jha is an agentic AI architect and the founder of GodEngine and AskShiva. Based in Palghar, Maharashtra, he builds AI-native systems designed to move beyond simple chat to automate complex business tasks and decision-making.

Who is Divya Prakash Jha?

Divya Prakash Jha is an agentic AI architect building decision intelligence and automated business assistants from Palghar, Maharashtra.

He is the founder of two specialized AI companies: GodEngine and AskShiva. His work focuses on creating systems that do more than just answer questions; he builds AI that can actually finish work and help humans make high-stakes decisions by providing structured reasoning.

  • Founder: Leads GodEngine and AskShiva.
  • Location: Based in Palghar, Maharashtra.
  • Specialization: Building agentic AI systems that move from providing answers to completing entire workflows.

From Trading Systems to AI

Before entering the world of AI, Divya spent a decade working in the high-stakes Gold and FX markets. Starting in 2016, he learned firsthand that markets are unforgiving and do not care about confidence—only whether a system holds up when conditions change. This lesson in discipline and automation shaped his approach to building AI.

During this time, he built over 500 automated trading systems, known as Expert Advisors, for MetaTrader 5. He served approximately 100 clients through specialized studios, including CandleForge, which focused on custom builds, and WICKFURY, which provided subscription-based software for trading desks.

This background in "small agents"—programs that read signals, decide, and act without human intervention—made the transition to Large Language Models (LLMs) feel natural. He applied the same logic: what does the system know, what is it allowed to do, and how do you verify it before it costs you money?

Decision Intelligence with GodEngine

Founded in 2026, GodEngine is a decision intelligence engine. While most AI tools are designed to produce fluent conversation, GodEngine is built to help people make difficult choices by providing structure and auditability.

When you ask a decision-based question—such as "Should we open a second outlet in Vasai this year, or invest in delivery first?"—the engine does not just give a chat response. It follows a "resolve pipeline" to break the problem down:

  1. Break it down: It extracts entities, facts, and cause-and-effect links from your question and places them into a clickable knowledge graph.
  2. Gather sources: It collects supporting data and attaches it directly to the facts.
  3. Build scenarios: It builds multiple plausible futures and probes them to see what would have to be true for each scenario to occur.
  4. Write the report: It streams a structured report showing reasoning, likely outcomes, risks, and citations.

This approach is designed for founders, analysts, and researchers who need to explain their decisions to others using a defensible, cited report rather than a single, unverified answer.

Automating Work with AskShiva

While GodEngine focuses on thinking, AskShiva focuses on doing. Founded in 2026, AskShiva is an AI assistant designed to bridge the gap between an AI providing an answer and a user completing a task.

For a small business owner, a "clever paragraph" isn't helpful if the work isn't done. AskShiva is built to handle the actual output. For example, if you ask it to "Make a GST bill for Sharma Traders, ₹18,400," it drafts the invoice with the correct tax lines and provides a downloadable file. If you then say, "Send it on WhatsApp, in Marathi," it handles the translation and prepares the message.

Key features of AskShiva include:

  • Memory: It maintains context across different conversations, so you don't have to re-explain who your customers are every time.
  • Voice: You can speak to the assistant in the browser and hear its reply.
  • Indian-Centric Workflow: It is built specifically for the Indian business environment, supporting Indian languages (like Marathi and Hindi) and essential documents like GST invoices.
  • Model Routing: It automatically picks the best AI model for the task and can switch models if one provider fails.

The Agentic Studio Model

Divya operates his AI development work through a unique "Agentic Studio" model. Instead of a large traditional agency, the studio consists of three human operators directing a fleet of fifteen AI agents. These agents are given names inspired by Sanskrit or Hindi to reflect their specific crafts, such as Vastu for architecture or Sutra for code.

This model allows a very small team to ship products at the scale of a much larger organization. The process follows a strict pattern: humans define the intent and the constraints, humans decide on the architecture to ensure leverage, and then the agents build the components in parallel.

To maintain control and reduce costs, the studio runs on its own private GPU and CPU servers using open models. This ensures that high-volume AI work does not depend entirely on paid, third-party APIs.

Building the BusinessOS

The ultimate vision for these technologies is BusinessOS, an AI-native operating system designed to replace the disconnected mess of modern business tools. Currently, many small and medium enterprises (SMEs) run their businesses using a patchwork of tools: leads in spreadsheets, customers on WhatsApp, HR in one app, and invoices in another. This makes it impossible to have a single "source of truth."

BusinessOS aims to provide a unified foundation for an entire company. By building the foundation first, every module built on top of it shares the same identity, permissions, and audit logs. This eliminates the security holes and duplicate data that occur when different software tools are bolted together.

The foundation of BusinessOS is already live in early release, providing secure sign-in, organization management, and a complete audit log. While the CRM is planned as the first business module, it is not yet built.

Comparison: Traditional Tools vs. BusinessOS

Feature Traditional SME Setup BusinessOS Approach
Data Flow Disconnected (Spreadsheets, WhatsApp, Email) Unified (Single database source of truth)
Security Fragmented permissions across apps One central permission engine
Visibility Hard to track who changed what Append-only, searchable audit logs
AI Integration AI only sees small pieces of data AI has a complete, unified context

By moving from "disconnected tools" to an "AI-native operating system," BusinessOS aims to give business owners a single, intelligent interface to run their entire operation.