Let's talk

Founder & Architect

Divyaprakash Jha

Building the future of agentic AI through GodEngine and AskShiva.

By Divyaprakash JhaUpdated 6 min readTopic: Founder

Photo for Divyaprakash Jha

Divyaprakash Jha is the founder of GodEngine and AskShiva, building agentic AI platforms and decision intelligence from Palghar, Maharashtra. He specializes in creating automated systems that move beyond simple chat to perform real-world business tasks and complex reasoning.

Who is Divyaprakash Jha?

Divyaprakash Jha is an architect of agentic AI systems with a deep background in algorithmic logic. His approach to AI is shaped by years of working in high-stakes environments where precision is the only thing that matters.

  • Founder of GodEngine and AskShiva: Leading two independent AI companies focused on decision intelligence and productivity.
  • Agentic AI Architect: Based in Palghar, Maharashtra, he builds systems that can plan, act, and verify their own work.
  • Trading Expert: He spent ten years in the Gold and FX markets, developing a disciplined approach to system design.
  • Proven Automation Track Record: He has built more than 500 automated trading systems (Expert Advisors) for MetaTrader 5.
  • Broad Client Experience: Through Dev.askshiva.ai, he has delivered websites, apps, and AI integrations for around a hundred clients, including clinics, retailers, exporters, schools, CA firms, and restaurants.

Decision Intelligence with GodEngine

Most AI tools are designed to answer questions. GodEngine is designed to help you make decisions. In high-stakes business environments, a simple chat reply often hides uncertainty behind confident-sounding prose. GodEngine is built to expose that uncertainty.

When you ask a question, the engine treats it as a model of the world. It doesn't just guess the next word; it maps out the logic. It is particularly useful for founders choosing between markets, analysts needing defensible reports, or operators weighing changes that affect multiple teams.

How it works

GodEngine uses a "resolve pipeline" that streams its reasoning to you in real-time. You don't have to wait for a wall of text; you watch the engine think.

  • Break it down: It extracts entities, facts, and cause-and-effect links from your question and places them in a clickable knowledge graph.
  • Gather sources: It collects supporting sources and attaches them to the specific facts they support, so you can audit the claims.
  • Build scenarios: It creates multiple plausible futures, probing what would have to be true for each scenario to happen and identifying what might break.
  • Write the report: It produces a structured report laying out options, likely outcomes, and risks, complete with citations.

By using a "figure check," the system compares numbers in its answers against computed facts to ensure accuracy. This makes it a tool for people who need to explain their decisions to others with confidence.

To learn more about the logic behind these systems, read the agentic architecture guide.

Productivity with AskShiva

For many small business owners, a "clever" AI paragraph is not helpful if it doesn't solve a practical problem. AskShiva is built to bridge the gap between a simple answer and finished work. It is designed specifically for the realities of Indian business, where workflows often move between WhatsApp, GST billing, and local languages.

AskShiva acts as an assistant that remembers your context. You don't have to re-explain who your customers are or what your business does every time you start a new chat. It maintains memory across conversations to provide seamless follow-ups.

Real-World Use Cases

Consider a typical business day for a merchant in Palghar or Mumbai:

  • Invoicing: You can say, "Make a GST bill for Sharma Traders, ₹18,400." AskShiva drafts the invoice with the correct tax lines and provides a downloadable file.
  • Communication: You can then say, "Send it on WhatsApp, in Marathi." It translates the message and prepares it for sending.
  • Follow-ups: Because it remembers the transaction, you can ask, "Remind me if they have not paid by Friday."
  • Data Conversion: You can ask it to turn a messy spreadsheet from last month into a clean, printable one-page price list.

To see the full range of capabilities, visit the AskShiva product page.

The Shiva Agents Platform

While a chatbot is a single loop, a company is a series of connected processes. The Shiva Agents platform is designed for the "second world"—the world where agents must run every morning, survive server restarts, and wait days for human approval before taking a risky action.

This platform allows you to run AI agents as an "always-on" company. It moves beyond simple demos and into production-grade automation that can handle complex, multi-step workflows.

Feature Description
Visual Workflows Draw workflows on a canvas (n8n-style) with over 20 node types.
Autonomy Levels Control agents from "Observe" to "Assist" to "Act" or "Autonomous."
Governance Every tool has a risk class (read, write, critical) gated by autonomy levels.
Open Protocols Supports MCP and A2A open agent protocols for interoperability.
Reliability Features token budgets, retries, timeouts, and checkpoints after every step.

The platform allows you to build an agent just like you would brief a new hire: defining its role, goal, backstory, style, and specific skills like research or data analysis.

BusinessOS: The AI-Native OS

Most small and medium enterprises (SMEs) suffer from "disconnected data." A business might run leads in a spreadsheet, customers on WhatsApp, HR in a separate app, and invoices in an email inbox. This makes it impossible to get a single view of the business.

BusinessOS is a multi-tenant, AI-native operating system designed to solve this fragmentation. Instead of bolting products together, it starts with a single, solid foundation for the entire organization.

The Foundation First Approach

BusinessOS provides a unified layer for identity, permissions, and data. This ensures that as you add more modules (like CRM), they all share the same security and information structure.

  • Secure Access: Uses passkeys, two-factor authentication, and SSO (Google/Microsoft).
  • Data Isolation: Uses database-level isolation (PostgreSQL Row-Level Security) so different organizations never see each other's data.
  • Unified Identity: One permission engine for all roles and access levels across the whole company.
  • Auditability: An append-only, searchable audit log that tracks every change and action.

By centralizing search, files, and notifications into one "event bus," BusinessOS allows AI agents to actually understand the full context of a business rather than just seeing one small piece of the puzzle.

Explore the architecture at BusinessOS.

How the Studio Operates

The development of these platforms follows a unique "Studio Model." Rather than a massive agency, the work is driven by a small, highly efficient team of three human operators directing a fleet of fifteen AI agents.

Each agent in the fleet is named after its specific craft, using Sanskrit or Hindi names to reflect its role. For example, Vastu handles architecture, Sutra handles code, and Ganit handles numbers.

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

From Trading Systems to AI

The logic used to build AI agents is deeply rooted in the discipline of algorithmic trading. In the markets, a system must be able to read signals, decide, act, and manage risk without a human needing to click a button. This is exactly what an "agentic" AI does.

Divyaprakash Jha's journey from building Expert Advisors for MetaTrader 5 to building AI platforms was a natural evolution. The core questions remain the same: What does the system know? What is it allowed to do? How do you check it before it costs you money?

This heritage of market discipline—where a single error can be costly—is what drives the focus on verification, audit logs, and structured reasoning in all of GodEngine's and AskShiva's products.

Frequently Asked Questions

What companies did Divyaprakash Jha found?
He founded GodEngine, a decision intelligence engine, and AskShiva, an AI assistant for business tasks.
Where is Divyaprakash Jha based?
He operates from Palghar, Maharashtra, India.
What is the difference between GodEngine and AskShiva?
GodEngine is for complex decision-making and scenario probing, while AskShiva is a task-oriented assistant that finishes work like invoices and documents.
How many AI agents does the studio use?
The studio uses a fleet of fifteen AI agents, each named in Sanskrit or Hindi, to assist three human operators.
Does AskShiva support Indian languages?
Yes, AskShiva is designed to handle Indian languages and business requirements, such as translating messages and creating GST invoices.

Questions people ask

What companies did Divyaprakash Jha found?

He founded GodEngine, a decision intelligence engine, and AskShiva, an AI assistant for business tasks.

Where is Divyaprakash Jha based?

He operates from Palghar, Maharashtra, India.

What is the difference between GodEngine and AskShiva?

GodEngine is for complex decision-making and scenario probing, while AskShiva is a task-oriented assistant that finishes work like invoices and documents.

How many AI agents does the studio use?

The studio uses a fleet of fifteen AI agents, each named in Sanskrit or Hindi, to assist three human operators.

Does AskShiva support Indian languages?

Yes, AskShiva is designed to handle Indian languages and business requirements, such as translating messages and creating GST invoices.