Consider the modest ambition of buying a snack on an aircraft. A crew member takes an order. A card is presented. Somewhere behind that small transaction sits a less modest problem: how do several sales devices agree about payments and stock when connectivity is unreliable? For Pratham Software, known as PSI, this is the kind of problem that makes an old application worth rebuilding.
- Builds and repairs the software other businesses sell and use.
- Specializes in B2B SaaS, cloud systems and enterprise modernization.
- The useful question: which daily bottleneck will the engineering remove?
A checkout without a connection
PSI describes a European travel-retail technology client whose platform handles onboard food, drinks and duty-free sales. The relationship had lasted more than six years. Its challenges included a Windows 8 desktop application, offline card processing, limited reporting and sales devices that lacked real-time synchronization.
The response combined a web-based point-of-sale application, offline payment integrations and an edge-cloud system that synchronized devices inside the aircraft. PSI reports a 29% increase in spending per passenger. That figure belongs to this particular project; the more portable lesson is the engineering choice. A system designed for a cabin has to accommodate the cabin’s conditions.
The company behind somebody else’s product
PSI began in Jaipur in 2000, founded by Puneet Mittal and Sumeeti Mittal. Its account of the early years gives an unusually specific supporting role to Pratibha Gupta: she introduced initial prospects, helped recruit staff and guided delivery. Before there was an international operation, there were introductions and people who could do the work.
Today, its services span software product engineering, enterprise applications, cloud migration, testing, data and AI. Customers include independent software vendors and enterprises that need help building or maintaining their own systems. Public customer reviews include sustainability collaboration platform OnePlanet and educational software business myKidzDay.
This puts PSI in the engineering-services market: customers buy development work and continuing technical capacity. A SaaS vendor can bring a product idea or an existing platform. An enterprise can bring applications that no longer cooperate. PSI’s proposed role stretches from architecture and implementation to testing and subsequent releases.
The SaaS work adds particular demands. PSI’s platform-engineering offering names multitenancy, frequent releases, scalability and configurability. In plain English, one product may serve many customer organizations, each expecting its own settings and dependable access. That requires decisions about how the application is divided, tested and operated. The alternative for a buyer is to develop those capabilities internally or hire another engineering provider. PSI competes on its ability to supply the necessary expertise across that continuing workload.
Forty percent fewer clicks
Another PSI case study concerns a European construction ERP provider. Its desktop system depended on local databases and manual synchronization. Remote access was limited; scaling and the interface were problems. The reason to change was already visible in the working day.
PSI describes replacing that arrangement with a multi-tenant architecture on Microsoft Azure, supporting desktop and web access across locations. Its reported result is pleasingly concrete: 40% fewer clicks per task. That measures interaction, rather than proving a corresponding reduction in elapsed time. Still, it gives a buyer something sharper to discuss than a promise of modernization.
PSI’s reported construction ERP result.
A usability measure, not a stopwatch.
A separate sustainability-platform rebuild involved performance trouble, an inefficient database, an aging technology stack and difficulty connecting to outside systems. PSI used React and Node.js, reorganized the architecture and moved the platform to Azure. The case illustrates a useful distinction: moving servers and fixing software are separate jobs. Here, the intervention included both.
The bill, and the work behind it
Custom engineering carries a project bill. PSI’s Clutch listing gives a $25,000-plus minimum and a $25-$49 hourly range. Those are directory indicators, not a quotation for a particular rebuild. In one historical customer interview, the buyer reported spending about £40,000 and expanding from one developer to four.
That customer also described opening business communication channels to the outside team. The practical inference is straightforward: an engineering partner needs access to the people who understand the product. A team receiving isolated tickets may struggle to make sensible trade-offs. Buyers can copy the habit of sharing context, while retaining a clear owner for priorities and acceptance decisions.
The sensible first purchase is therefore a bounded piece of work with an observable result. Ask how success will be measured, which integrations are included and who maintains the result after delivery. If a standard software package already meets the need, a custom engineering engagement may add cost and coordination without sufficient benefit. That judgment belongs ahead of the technology decision.
A training library gets a second life
PSI also offers a more packaged proposition. xTract2AI extracts content from SCORM, a common format for online training packages, and turns it into structured material intended for AI applications. The problem is recognizable: a company has already paid to create courses, but their packaging makes the knowledge awkward to reuse.
Its published pay-per-use option costs $30 per course, subject to package-size and media limits. Larger bundles and a custom enterprise option are also listed. This offers a narrowly defined starting point for learning teams with existing archives. Its value depends on having relevant source material and a useful destination for the extracted content.
Alongside it, PSI’s homepage promotes CodePulse AI, described as turning test plans into automation. Both products sit close to the firm’s services: recovering useful work from an existing asset, or reducing a repetitive engineering handoff.
The opportunity you let go
The company’s culture page contains a demanding staffing principle. It says PSI will hire suitable people even when the alternative is losing a business opportunity. It also states:
“We will never oversell our services.”
PSI’s published culture principles
These are stated commitments. For a customer, their value lies in whether they shape an actual scope discussion. Can the team explain the limitation, identify the right skills and set a measurable goal?
PSI’s projects suggest a useful buying discipline: begin with the transaction, the synchronization problem or the unnecessary clicks. Choose the work around that constraint. A modern stack earns its keep when the person using it can get on with the job.