Intelligence Digitally Delivered

AI Labs@Mechsoft

AI before AI! A brief history

stone
stone

In 1996 - when the AI lingo was not so hot,

Mechsoft was formed with a tagline Architects of Virtual Universe

And the Vision was defined as –

|| Vision ||

Take Information Technology close to Human Intellect.

Adding Intellect, Transparency & Integrity to any aspect of

life, Business, Education or Governance!.

To create a Virtual universe where the world is a happier place!.

With our limited resources we started on this journey.

First attempt was to create Interactive education.

The idea was to train the engine on large information, and it would be an intelligent educator, that would understand your specific needs, strengths – will be able to explain in various ways, like giving example, analogies, draw a diagram to explain etc. It would be able to judge your level of understanding and guide you accordingly (example – it would say – you have not understood basics, please study pointers first and then come back to this chapter)

stone
stone

Second attempt was an application creator. Simple – you interact with it, it would ask questions to clarify your requirements, scan your printed reports, and based on that, understand name, date, money etc, again ask a few questions and understand the data structure needed, give you a few interface ideas, and based on your selection – create an application.

But there were 2 major issues.

One – People found it impossible to achieve. (in one event – where we were selected among top innovators a professor was guiding innovators – told me, ‘many have tried doing automated coding, but it is not possible’)

Second – We quickly ran out of funds. Had to give these ideas to almost freshers, who struggled to even understand the concepts.

The tasks were not impossible, but were pretty complex, to write the whole AI engine – every part by ourselves.

stone
stone

First attempt was to create Interactive education.

The idea was to train the engine on large information, and it would be an intelligent educator, that would understand your specific needs, strengths – will be able to explain in various ways, like giving example, analogies, draw a diagram to explain etc. It would be able to judge your level of understanding and guide you accordingly (example – it would say – you have not understood basics, please study pointers first and then come back to this chapter)

stone
stone
stone

Second attempt was an application creator. Simple – you interact with it, it would ask questions to clarify your requirements, scan your printed reports, and based on that, understand name, date, money etc, again ask a few questions and understand the data structure needed, give you a few interface ideas, and based on your selection – create an application.

stone

But there were 2 major issues.

One – People found it impossible to achieve. (in one event – where we were selected among top innovators a professor was guiding innovators – told me, ‘many have tried doing automated coding, but it is not possible’)

Second – We quickly ran out of funds. Had to give these ideas to almost freshers, who struggled to even understand the concepts.

The tasks were not impossible, but were pretty complex, to write the whole AI engine – every part by ourselves.

And AI arrived in main stream.

What looked impossible of science fiction few years back, is now easily available.

Some of the projects we had in mind, would be easier to develop mow.

Half of our vision is more or less achieved by the technology world. So do we have a role?

Close to human intellect – is achieved but what about Adding integrity, equality? People are worried about the unethical usage of AI.

We are now working on AI to achieve this – to help society – mankind, to bring more equality and honesty to the whole system.

How do we achieve these ethical advances in AI?

stone

With advent of AI, many businesses/ streams are likely to be disrupted.

People are saying Go AI or go bust!

At Mechsoft we are working on 2 very interesting streams.

stone

Work on AI Technology

Use of various AI engines, integrate them to help businesses to grow/ save costs/ increase efficiency. Work on our own AI engines and offer them to public for integration with their products/ services.

As we explain in our article You need business understanding and natural intellect here are Case studies of actual application of AI tools for significant advantages in business processes.

Case Studies

1. Large & Complex Support system: Transformation using AI and social media!

Rule based billing

Struggling with cumbersome support processes across borders, our retail client faced many challenges deploying applications efficiently. Human support was time-consuming and error-prone, exacerbating operational bottlenecks. Existing chatbot solutions lacked true AI capabilities, relying on rigid, pre-designed flows similar to IVR systems.

Addressing to these limitations we created fully working prototype, our TRUE AI-powered chatbot, seamlessly integrated with WhatsApp. It leverages Rasa NLU's machine learning algorithms, we're fine-tuning our prototype to analyze user input, predict intents, and identify entities with precision. With a robust training dataset tailored to our client's support needs, our NLP engine is poised to evolve into a responsive, intuitive solution.

2. Beyond Integration: AI for Image generation for promotions!
Rule based billing

In collaboration with a US client, we implemented OpenAI's DALL-E API to revolutionize their promotional image generation process. Our engagement went beyond mere tool integration. By analyzing historical data, we tailored the API's input parameters to better suit the end-users' needs, integrating these insights directly into the requirements. This not only streamlined the image generation process but also significantly reduced both time and costs for our client. Our focused enhancements enabled the client to achieve their desired designs swiftly and more efficiently, optimizing AI API usage to minimize operational costs.

Work on AI Proof Technologies

What is AI proof technology? These are ideas and concepts, that are not likely to be disrupted by AI. We are sure, if any investor is presented with an idea, the first thing that would come to their mind would be – ‘Would some new AI stuff be able to disrupt this?’ AI proof technologies are human centric, will depend on, be based on human inputs and thus are AI-Proof Technologies.

Interestingly AI proof – which can be a great opportunity for a stable business case – is likely to do good for humanity as well, as it is human centric. Here is a high level info about some projects we are working on, in the area of AI proof – Human Based Tech.

1. CloseCircle:

Rule based billing

A compassionate innovative concept where people can support each other – the most natural way of social security – and create a social media platform - something that will have human touch. We will use AI for making this better – but the concept is such that humans will never be replaced by AI!

2. OpaqueDoor:
Rule based billing

Again a human centric concept to disrupt the job search market which is completely different from the current Job portals, and something that enforces much better accuracy and integrity.

Will AI help my business? How? First read this before spending a fortune!

Real AI vs Fake AI.

You need business understanding and natural intellect