Let's talk

Founder & Architect

Divyaprakash Jha in Palghar

An agentic AI architect building the future of decision intelligence and business automation from Palghar, Maharashtra.

By Divyaprakash JhaUpdated 4 min readTopic: Founder

Photo for Divyaprakash Jha in Palghar

Divyaprakash Jha is an agentic AI architect based in Palghar, Maharashtra, where he operates a specialized AI development studio. He is the founder of GodEngine and AskShiva, focusing on building decision intelligence and business automation tools.

Based in Palghar, Maharashtra

Operating from Palghar, Maharashtra, Divyaprakash Jha builds advanced AI systems designed for real-world utility. His work is not just about writing code, but about building digital structures that can function like a professional team.

The studio operates with a unique hybrid model of human intelligence and machine efficiency. It consists of three human operators directing a fleet of fifteen AI agents. This setup allows the studio to ship complex software projects that would typically require a much larger agency.

To maintain independence and control over costs, the team uses local GPU and CPU servers to run open models for bulk AI work. This infrastructure allows them to handle large-scale AI tasks without being entirely dependent on expensive, third-party paid APIs.

Founder of Two AI Companies

After a decade of building systems for the financial markets, Divyaprakash Jha transitioned into full-time AI development. In April 2026, he went fully independent to focus entirely on the architecture of agentic AI.

From this foundation, he founded two distinct companies, each solving a different side of the intelligence problem:

  • GodEngine: A decision intelligence engine designed to help leaders make high-stakes choices by showing the reasoning behind every conclusion.
  • AskShiva: An AI assistant built to bridge the gap between a simple chat response and a completed business task, specifically optimized for Indian languages and workflows.

While GodEngine focuses on the "why" and the "what if" of complex decisions, AskShiva focuses on the "do"—handling the actual execution of daily business chores.

From Financial Markets to AI

The precision required in AI architecture was forged in the high-stakes environment of the financial markets. Starting in 2016, Divyaprakash Jha spent ten years working in the Gold and FX markets. He learned early on that markets are an unforgiving teacher; they do not care about confidence, only whether a system holds up when conditions change.

This decade of experience in trading led to the creation of over 500 automated trading systems, known as Expert Advisors (EAs), for MetaTrader 5. These systems were designed to execute rules with strict discipline, acting as small, autonomous agents that read signals and manage risk without human intervention.

From this work, two specialized studios emerged to serve the trading community:

  1. CandleForge: A studio dedicated to custom Expert Advisors and full-stack builds for traders.
  2. WICKFURY: A studio that licenses MT5 Expert Advisors on a subscription basis and builds specialized software for professional trading desks.

Agentic Architecture & Design

Divyaprakash Jha does not build simple chatbots. Instead, he builds "agentic" platforms. In a standard chatbot, you ask a question and get a text answer. In an agentic system, the AI follows a "plan, act, verify" loop. It creates a plan, uses tools to perform the work, and then checks its own work to ensure accuracy.

This philosophy is reflected in the way his agents are named and organized. The studio uses a fleet of AI agents with Sanskrit or Hindi names to represent their specific crafts:

Agent Name Specialization / Craft
Vastu Architecture
Akshar Lettering / Text
Vani Words / Speech
Ganit Numbers / Calculation
Sutra Code / Logic
Parakh Testing / Verification

The ultimate goal of this architecture is "decision intelligence." Rather than just providing a fluent-sounding sentence, the systems provide structured reasoning. This allows a business owner to audit the AI, seeing exactly which facts led to a specific recommendation.

Empowering Small Businesses

Many AI tools are designed for developers or large tech companies, often ignoring the practical needs of small business owners. Divyaprakash Jha’s work focuses on the "gap between an answer and a done task."

For example, a small business owner in India doesn't just need a paragraph explaining how to make a GST invoice; they need the actual invoice generated, formatted correctly, and sent to the customer via WhatsApp in a language they understand, such as Marathi or Hindi.

This vision is being realized through two main avenues:

  • AskShiva: An assistant that handles multi-step tasks like creating billing, translating customer messages, and managing follow-ups.
  • BusinessOS: An AI-native operating system designed to act as the central nervous system for an entire organization.

Currently, many small and medium enterprises (SMEs) run their businesses using disconnected tools: spreadsheets for leads, WhatsApp for customers, and email for invoices. BusinessOS aims to bring all of these into one secure, integrated foundation where data flows seamlessly between departments.

The Engineering Approach

The engineering behind these platforms is built for durability and auditability. Unlike many AI applications that rely on a single "black box" model, this architecture is designed to be transparent and resilient.

Key technical focuses include:

Auditability and Verification

Every action taken by an AI agent is recorded. In systems like Shiva Agents, the studio uses hash-chained records of every action. This creates a permanent, unchangeable audit trail, so a manager can see exactly what an agent did, why it did it, and whether it followed the rules.

Resilience and "Time Travel"

Real-world work doesn't always happen in one go. An agent might need to wait two days for a manager to approve a refund or a large transaction. The engineering is designed so that agents can survive server restarts and handle multi-day approval processes without losing their place in a task.

Furthermore, the system allows for "time travel"—the ability to fork a workflow from an earlier step. This allows developers and business owners to test different paths or "what-if" scenarios without breaking the main process.

Infrastructure Independence

By running an independent infrastructure with open models, the studio avoids the risks associated with relying solely on paid, third-party APIs. This ensures that bulk AI work remains cost-effective and that the business is not at the mercy of a single provider's uptime or pricing changes.

Questions people ask

Where is Divyaprakash Jha located?

He is based in Palghar, Maharashtra, where he operates his AI development studio.

What companies did Divyaprakash Jha found?

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

What is his background in trading?

He has ten years of experience in Gold and FX markets, having built over 500 automated trading systems for MT5.

What is the purpose of AskShiva?

AskShiva is an AI assistant that handles multi-step tasks like creating GST invoices, translating messages into Marathi, and managing follow-ups.

How does his AI studio operate?

The studio uses a hybrid model of three human operators directing a fleet of fifteen specialized AI agents.