Skip to content
Founded 2026 — taking new engagements

Ship the hard parts.

Codeaex is an AI and full stack development studio. We build the hard parts: the machine learning that makes a product intelligent and the software around it that puts it in front of users — then hand you something your own team can run.

Start a projectSee the architecture
both halves, one team
AI + Web

both halves, one team

between working demos
2 weeks

between working demos

code and cloud stay yours
100%

code and cloud stay yours

// Blue/green rollout with automatic rollback
export async function release(svc: Service) {
  const next = await svc.provision({ replicas: 12 })
  await next.warm()
  if (await healthy(next, { p99: 80 })) {
    return svc.cutover(next)
  }
  await svc.rollback()
  throw new ReleaseError("p99 regression")
}
typecheck passing
zsh — codeaex
$

Built with

Technologies we work in: TypeScript, JavaScript, Python, React, Next.js, Node.js, Express, FastAPI, Django, PostgreSQL, pgvector, Prisma, MongoDB, Redis, LangChain, Claude API, Hugging Face, PyTorch, scikit-learn, Pandas, Docker, Terraform, GitHub Actions, AWS, Google Cloud, Vercel, OpenTelemetry, Prometheus, Grafana, Sentry
Services

Six practices, one team.

Most engagements draw on three or four of these at once — which is precisely why they live under one roof.

All services

AI & LLM applications

Chatbots, retrieval, and agents that answer from your own content — shipped with evaluation and cost ceilings.

  • RAG & hybrid search
  • Agent orchestration
  • Eval + guardrails

Python · LangChain · pgvector

Machine learning & analytics

Forecasting, classification, and recommendation models that stay accurate after handover.

  • Feature engineering
  • Model selection & tuning
  • Scheduled retraining

Python · scikit-learn · PyTorch

Full stack product engineering

Typed end-to-end interfaces, design systems, and front ends that stay fast as they grow.

  • Design systems
  • Edge rendering
  • Accessibility to WCAG AA

TypeScript · React · Next.js

Backend, APIs & integrations

Services, authentication, billing, and the third-party plumbing a product needs to actually transact.

  • REST & typed APIs
  • Auth and permissions
  • Payments & webhooks

Node.js · FastAPI · PostgreSQL

Data pipelines & storage

Ingest, clean, and model the data that your product and your models both depend on.

  • ETL & document ingest
  • Schema design
  • Vector + relational stores

PostgreSQL · pgvector · Pandas

Cloud, deployment & support

Infrastructure as code, CI that ships in minutes, and someone who answers after launch.

  • IaC from day one
  • Progressive delivery
  • Post-launch support

Docker · AWS · Vercel

Reference architecture

This is what we hand over.

A composite of the systems we build: typed at every boundary, observable end to end, and deployable by your team without us in the room.

system.topologylive traffic
EDGEAPPLICATIONDATA & MODELSOPERATIONSClientsWeb · Mobile · Partner APIEdge / CDNCache · WAF · TLSAPI GatewayAuthN/Z · Rate limitServicesNode · FastAPI · typedWorkersAsync jobs · retriesOLTPPostgreSQL · replicasModel layerInference · eval · fallbackVector storepgvector · citationsTelemetryOpenTelemetry · tracesCI / CDIaC · progressive delivery

Scroll to pan the diagram

Typed at every boundary

Schemas are versioned and generated into clients. No untyped JSON crossing a service line, and no silent contract drift between teams.

Grounded, not guessed

Anything a model outputs can be traced to the source that produced it. Where there is no grounding, the system escalates to a person instead of inventing an answer.

Reproducible from zero

Every environment can be rebuilt from a clean cloud account with infrastructure as code. Nothing important exists only in someone's console history.

Reversible by default

Progressive delivery, feature flags, and a rollback path that is tested rather than assumed. A bad release is an inconvenience, not an incident.

Selected work

Systems that went to production.

Three engagements, described the way an engineer would describe them: the problem, the approach, and the numbers afterwards.

All case studies
CodeaexAccreditation / Multi-tenant SaaS2026

AMS — a multi-tenant accreditation platform

The full accreditation lifecycle in one system: company onboarding, applications, assessments, assessor scheduling, certificates and invoicing — running in production.

  • FastAPI
  • Next.js
  • PostgreSQL
  • Celery
  • Playwright
Read the case study
Invite-only provisioning
Tenancy
Server-rendered PDFs
Documents
Redis + Celery
Background work

Illustrative shape

CodeaexAI / Conversational SaaS2026

ChatDesk — AI chatbots for any business

One platform to create, sell and manage AI assistants for many businesses — each one answering customers on the website and on WhatsApp from that business's own documents.

  • Next.js
  • TypeScript
  • PostgreSQL
  • pgvector
  • Gemini / Claude
Read the case study
Website + WhatsApp
Channels
English, Urdu, Roman Urdu
Languages
Login per business
Tenancy

Illustrative shape

CodeaexSME / Business management SaaS2026

FMS — sales, stock and money for small businesses

Business management software for SMEs: invoices, stock, customers, suppliers, credit (udhaar) and profit in one multi-tenant system.

  • FastAPI
  • React
  • TypeScript
  • PostgreSQL
  • Docker
Read the case study
Sales, stock, customers, suppliers
Modules
Udhaar + part payments
Credit
Append-only
Ledger

Illustrative shape

CodeaexAutomation / WhatsApp Business2026

WhatsApp automation that replies, books and captures leads

An AI assistant on a business's own WhatsApp number, connected through the official Meta Cloud API — answering, taking bookings and orders, and saving every customer as a lead.

  • WhatsApp Cloud API
  • Next.js
  • TypeScript
  • PostgreSQL
Read the case study
Official Meta Cloud API
Integration
Signature-verified
Webhooks
Handled exactly once
Duplicates

Illustrative shape

CodeaexAI / Retrieval-augmented generation2026

A RAG engine that answers from your own documents

Upload PDFs, Word files and spreadsheets, or import a website — the assistant answers from that content and nothing else.

  • TypeScript
  • PostgreSQL
  • pgvector
  • Gemini embeddings
  • Next.js
Read the case study
PDF, Word, CSV, websites
Sources
Hybrid keyword + vector
Search
Top 6 sections per question
Context

Illustrative shape

Reference buildSaaS / Product2026

A SaaS foundation that survives its own launch

Multi-tenant workspaces, roles and permissions, subscription billing, and a dashboard — the parts every SaaS needs before it can charge anybody.

  • Next.js
  • TypeScript
  • Node.js
  • PostgreSQL
  • Stripe
Read the case study
Row-level isolation
Tenancy
Stripe + webhooks
Billing
Roles & permissions
Access model

Illustrative shape

Reference buildMachine learning / Operations2026

Forecasting that still works after handover

A demand model trained on your history, served behind an API, retrained on a schedule, and explained well enough that a planner will actually use it.

  • Python
  • scikit-learn
  • XGBoost
  • FastAPI
  • React
Read the case study
Held-out period
Validation
Scheduled + tracked
Retraining
Per-prediction drivers
Explainability

Illustrative shape

Engineering standards

The targets we build to.

Every system we ship is instrumented against these from the first deploy, reporting into a dashboard you own. These are the targets we agree to before a build starts.

p95 latency

target

0ms

Uptime target

SLO

0.00%

Change lead time

commit to production

0min

Release gates

tests + evals, every deploy

0%

Process

Four stages, no surprises.

The same sequence whether the engagement runs six weeks or a year. Predictability is a feature.

  1. 01

    Technical discovery

    Week 1

    We read the code, trace the hot paths, and talk to the people who get called when it breaks. You get a written assessment of what is actually in the way — including the parts you did not ask about.

    You receive: Assessment + risk register

  2. 02

    Architecture & plan

    Week 2

    Target architecture, build sequence, delivery plan, and a fixed commercial shape. Enough detail to decide with your eyes open, and cheap enough to walk away from.

    You receive: Architecture notes + sequenced roadmap

  3. 03

    Build in increments

    Weeks 3–14

    Two-week increments, each deployed behind a flag and demonstrable in your environment. You see running software, not slides about running software.

    You receive: Deployed increments + CI

  4. 04

    Handover & support

    Ongoing

    Runbooks, architecture decision records, and a walkthrough with the engineers who will own it next. After launch we stay available on an agreed support window, sized to how much of it you want to run yourself.

    You receive: Runbooks + SLOs + support window

Engagements

Tell us what is breaking.

Send the problem, not a polished brief. An engineer reads every enquiry and replies within two working days with a real technical opinion — including when the honest answer is that you do not need us.

Start a projectcorp@codeaex.com
  • Reply from an engineer, not a salesperson
  • Fixed scope and price before any build starts
  • Your code, your cloud, your repo — from day one