All articles
Software DevelopmentAI DevelopmentGuides

6 Projects We're Proud Of, and the Hard Part of Each

Six real products our team designed and built, the hardest problem in each one, and how we run a four-week MVP in practice. No theory, only delivered work.

By the Clarigo Editorial Team · November 17, 2026 7 min read

Share
Short answer: The six projects below are a social network, a mental wellness app, an AI companion, a clinic front-office product, an electronic health record and a factory platform. In every one, the hard part was not writing code. It was deciding what the product must never do, and keeping it simple enough for people to use every day.

Most agency articles describe how software should be built. This one describes software we did build. Two of the projects are client products, and four are products our own team conceived, built and runs. For each, we explain what it is, what made it hard, and the decision that solved it. Every project links to its full case study so you can check the details.

The six projects at a glance

ProjectTypeIndustryWhat made it hard
LivSocialClient productSocial mediaA full social platform on a startup budget and timeline
SoothfyClient productMental wellnessWellness apps are abandoned within days
VumiOur own productAI companionWarm and personal, without being addictive or dishonest
Clarigo Clinic OSOur own productHealthcareAutomating the front desk without making medical decisions
Clarigo EHROur own productHealthcareOne system for the whole practice, with data that stays exportable
Clarigo Factory OSOur own productManufacturingAI on a production line that cannot depend on the internet

1. LivSocial: a new social network, built by one team

LivSocial is a social media platform founded by Laurie Edwards, built around real people and healthier conversation instead of forced content and endless ads. We designed and built the mobile app, the website and an AI tool inside the platform, and we have worked with the client since January 2025.

The hard part. A social network has to feel familiar from the first tap, or people go back to the networks they already use. LivSocial needed sign-up, posting, communities, business pages and a public discussion space called The SpeakEasy, all from a small team on a startup budget and timeline.

What solved it. One in-house team handled design, web, mobile, AI and cloud, so nothing was lost between vendors. The web and mobile apps share the same experience, and the same team that launched the beta keeps adding features. The platform is live in beta for real users. Read the full LivSocial case study.

2. Soothfy: a wellness app people keep using

Soothfy is a mental wellness app and website with a personalised 30-day journey for ADHD, anxiety, depression and sleep.

The hard part. Wellness apps often overwhelm people with choices, and most are abandoned within days. The product had to be simple enough to follow daily, personal enough to feel relevant, and fresh enough to bring people back, while treating a sensitive subject with care and privacy.

What solved it. A structure we call the Anchor + Novelty System: three short daily anchor routines for consistency, plus fresh AI-suggested activities for variety. The website also offers free, private two-minute assessments with no login, so people can start before they commit to anything. Both the website and the mobile app are live. Read the Soothfy case study.

3. Vumi: an AI companion that stays honest

Vumi is an AI best friend you can call, chat with and share your day with. It is a product our team conceived, designed, built and runs, and it is opening through a waitlist.

The hard part. An AI companion has to feel warm, talk out loud, remember what matters and speak people's languages. It also has to stay honest about what it is, protect privacy, and never try to keep people hooked. Those goals pull against each other.

What solved it. We wrote the limits into the product. Nothing is sent to the AI services behind Vumi until the person allows it, and memories and accounts can be deleted at any time. There are no ads and no endless scrolling, Vumi nudges people back towards real friends, and it shows crisis helplines instead of making therapy claims. Around those limits we built real-time voice calls, long-term memory, proactive check-ins and support for 27 languages. Read the Vumi case study.

4. Clarigo Clinic OS: automation that knows where to stop

Clarigo Clinic OS handles front-office work for specialty clinics: phone calls, intake forms, insurance verification, prior authorizations, denied claims and reminders. It grew out of our healthcare projects, where the same front-desk problems kept appearing.

The hard part. Healthcare automation has to be careful. The product had to take real work off the front desk while never replacing the clinic's EHR, never making a medical or financial decision on its own, and protecting patient data at every step.

What solved it. Three rules. A person approves anything medical or financial. Every action is logged in a full audit trail, with a BAA in place before any patient data is touched. And the product is split into six separate parts, so a clinic can start with one, usually the phone assistant, and add more once it trusts the system. It sits on top of the clinic's existing EHR, booking system and phone number, with no data migration. Read the Clinic OS case study.

5. Clarigo EHR: AI that removes typing, not judgement

Clarigo EHR is a cloud-based electronic health record and practice management platform: charting, scheduling, a patient portal, telehealth, billing, e-prescribing and reporting in one system.

The hard part. Many practices stitch together separate tools, and clinicians spend hours typing. We wanted AI to remove that busywork at every step while every clinical decision stays with a person, and we wanted the data to be secure, interoperable and always exportable.

What solved it. The AI scribe drafts the note and the clinician reviews and signs it. The platform supports HL7 and FHIR so it can exchange data with labs, pharmacies and clearinghouses, and a practice can export its data in full. Read the Clarigo EHR case study.

6. Clarigo Factory OS: AI that keeps working offline

Clarigo Factory OS brings predictive maintenance, vision quality inspection, live production metrics and safety monitoring into one platform for manufacturers.

The hard part. A production line cannot depend on the internet, and nobody wants software taking control of machinery. The platform had to work with existing PLCs, SCADA and MES, keep data inside the plant, and never change a machine setting without a person approving it.

What solved it. Processing runs at the edge, using the protocols plants already have (OPC UA, Modbus and MQTT), with air-gapped options for sites that need them. Process changes are suggested and a person approves them. The platform is designed to be piloted on one line, prove its value there, and then roll out across the plant. Read the Factory OS case study.

What the six have in common

Looking across these projects, the same four decisions appear every time:

  • Limits come first. Each product has a short list of things it must never do: make a clinical decision, hold data hostage, take over a machine, keep someone scrolling. Writing that list early makes every later decision easier.
  • Start with one part. Clinic OS starts with the phone. Factory OS starts with one line. Soothfy starts with a two-minute assessment. A small first step earns the trust that a big launch cannot.
  • Sit on top of what exists. Clinics keep their EHR, and factories keep their PLCs and SCADA. Replacing systems people depend on is the slowest and riskiest way to deliver value.
  • One team, start to finish. Design, engineering, AI and cloud sit in the same in-house team, and the same people stay after launch.

How this works in practice: our four-week MVP

The same habits shape how we build a first version for a client. Our MVP programme runs in roughly 25 to 30 days:

WeekWhat happensWhat you have at the end
Week 1: discovery and designScope workshop, user flows, wireframes, a clickable prototype and the technical planAn approved design and build plan
Week 2: core buildCloud architecture, database, sign-in and the core featuresA working app in staging
Week 3: AI and integrationsModel integration, automation workflows, payments and third-party systemsAI features live in staging
Week 4: test, polish and launchQuality, performance and security checks, deployment, documentation and handoverThe product live in production

You own the full source code, it is deployed on your own cloud, and the same engineers are available afterwards to keep improving it.

What it costs

Cost depends on scope, so we do not publish a single price. Our senior engineers typically cost US$25 to US$49 per hour, and the fastest way to get a realistic range for your own idea is our free app cost calculator, which estimates hours, timeline and team from the features you choose. For a breakdown of where the money goes, see how much it costs to build an app in India.

About Clarigo Infotech

Clarigo Infotech is an AI-first software and app development company based in Indore, India. Since 2016 our in-house team has delivered 600+ projects for clients in 20+ countries. You can browse more of our work in the portfolio, or tell us about your project and a senior engineer will reply with a scoped plan.

Frequently asked questions

Are these real projects?

+

Yes. LivSocial and Soothfy are client products, and Vumi, Clarigo Clinic OS, Clarigo EHR and Clarigo Factory OS are products Clarigo Infotech designed and built itself. Each one has a full case study on our website.

Which of these projects are live?

+

LivSocial is live in beta for real users, and Soothfy has a live website and mobile app. Vumi has a live website with a waitlist, and Clarigo Clinic OS, Clarigo EHR and Clarigo Factory OS are available to clinics, practices and plants.

How long does it take Clarigo to build an MVP?

+

Our MVP programme takes roughly 25 to 30 days: one week of discovery and design, one week of core build, one week of AI and integrations, and a final week of testing, polish and launch.

Who owns the code when Clarigo builds a product?

+

You do. Clients receive full source-code ownership, with the product deployed on their own cloud and handed over with documentation.

How much does a project like these cost?

+

It depends on the scope. Senior engineers typically cost US$25 to US$49 per hour, and our free app cost calculator gives an estimated range, timeline and team for the features you choose.

From our own work

Projects our team has delivered that relate to this article.

About the author

Clarigo Editorial Team

Written by the team at Clarigo Infotech, an AI-first software and app development company in Indore, India. Since 2016 our in-house team has delivered 600+ projects for clients in 20+ countries, and our articles draw on that work. See our case studies and portfolio.

Found this useful? Share it with your team.

Planning something like this?

Get a free consultation. A senior engineer replies within one working day with a scoped plan.

How should we contact you?

No obligation. We reply within one working day.

Keep reading