VixitAI Studios
VixitAI Studios

Software that survives its first hundred customers.

Most agencies can build you a demo. The hard part is everything after: tenant isolation that actually isolates, billing that reconciles, counters that do not collide under load, and a deployment you can repeat at 2am. That is the part we are good at.

Work

Products running in production, not concept decks.

Read the case studies
01
Healthcare SaaS

Clinic AI

A hospital management system in daily clinical use — appointments, EMR, lab, pharmacy, imaging and billing, with per-tenant isolation and audit logging throughout.

Multi-tenantEMRBillingRazorpay
Visit the product
02
Sports analytics

CricAI

Live scores and match analytics with a server-rendered content layer built for search.

Live dataSEOCMS
Visit the product
03
Publishing

Vixit News

A news surface with AI summarisation and regional language support, tuned for Core Web Vitals.

EditorialSummarisationPerformance
Visit the product
04
Platform

VixitAI

The umbrella platform the rest of the portfolio hangs off — shared auth, shared billing, and the deployment discipline that keeps a dozen services on one box behaving.

PlatformAuthInfrastructure
Visit the product
Capabilities

What we actually do, in the order it usually matters.

01

Backend and data

Node and Express APIs over tenant-isolated PostgreSQL, with the isolation boundary chosen deliberately rather than by default. Redis for queues and cache. Atomic, race-tested counters behind every human-readable identifier, because SELECT MAX + 1 works right up until two requests land in the same millisecond.

  • PostgreSQL + Redis
  • Role-based access control
  • Audit logging
  • Race-tested ID counters
02

AI that is load-bearing

Retrieval-augmented generation with hybrid vector and full-text search, reciprocal rank fusion and cross-encoder reranking. Multi-provider fallback so one slow vendor does not stall a request, per-tenant token metering so inference cost never surprises anyone, and evaluation harnesses so a prompt change cannot quietly regress an answer.

  • RAG pipelines
  • Provider failover
  • Token metering
  • Evaluation harnesses
03

Shipping and staying up

PM2 cluster mode behind nginx, zero-downtime reloads, encrypted off-site backups, and health checks that alert a human. Deployment is a script that runs the same way every time, not a sequence someone remembers.

  • PM2 + nginx
  • Zero-downtime deploys
  • Encrypted backups
  • Monitoring and alerts
Measured

Numbers we can show you the run for.

No compliance badges and no uptime guarantee — those mean an audit or an SLA with money behind it, and we do not have either yet. These are Lighthouse runs.

30 to 97
Mobile Lighthouse, this site

Rebuilt to remove render-blocking third parties and self-host every font.

74 to 89
Mobile Lighthouse, Rentlelo

A live marketplace, improved without changing what the product does.

0.26 to 0.019
Cumulative Layout Shift, 3D build

Metric-matched font fallbacks and reserved media boxes on a WebGL page.

How an engagement runs

We start with a written architecture note — data model, tenancy boundary, auth model, and the three things most likely to bite. It is short, and it is the most valuable document in the project, because it is where disagreements surface while they are still cheap.

Then we build in vertical slices you can actually use, not horizontal layers you have to take on faith. You see a working thing early and you keep seeing working things.

Tell us what you are building.

A short description of the problem is enough to start. If it is not something we should take on, we will say so.