Ahtesham Ahmad

Open to opportunities

I build AI systems that run businesses on their own.

5+ years shipping production AI. From email-to-task pipelines and multi-agent orchestration to open-source code review tools, I handle the full stack so you don't need a team.

How I architect every system

IngestClassifyOrchestrateReviewExecuteObserve

About

Most AI projects break in production.
I make sure mine don't.

The gap between an AI demo and a production system is enormous. Hallucinations slip through, there's no human oversight, and the whole thing is held together by a single API call. I've spent five years closing that gap.

I work where AI orchestration meets systems engineering: multi-agent architectures with human approval gates, workflow engines that connect LLM reasoning to real actions like calendar bookings and invoices, and full-stack products shipped from schema design to polished UI.

Design the system first. Then let AI agents handle the repetitive work while humans stay in control.

250+

Projects Shipped

5+

Years Building

98%

Client Satisfaction

<1s

CodeLens Reviews

Experience

Where I've shipped production systems.

Lead AI Developer

More Life Hospitality GmbH

Sep 2025 – Present

  • Building OpenEvent, an AI-powered event management SaaS handling real client workloads
  • Designed the full agentic pipeline: emails come in, AI classifies them, humans approve actions, workflows execute automatically
  • Architected the data model and AI orchestration layer on Supabase Edge Functions with OpenAI and pgvector
  • Shipped the React frontend: Task Inbox, Workflow Runner, Approval Flows, Marketing Hub, Admin Dashboard
  • Integrated Stripe billing, Google/Outlook calendars, row-level security, and CI/CD with GitHub Actions
ReactTypeScriptSupabaseOpenAIpgvectorStripeDocker

Director of IT & R&D

Rouelite Techno Pvt. Ltd.

2022 – 2024

  • Built a CRM from scratch that cut the solar project lifecycle from 14 days to 5
  • Introduced AI into daily operations, reducing manual data entry by 70%
  • Ran an agile team of 6 engineers, shipping quarterly releases
  • Replaced 3 legacy spreadsheet processes with React + Supabase internal tools
ReactSupabasePostgreSQLTailwind CSS

Freelance AI Consultant

Fiverr, Upwork & Wadware House

2019 – Present

  • Shipped 250+ projects: prompt engineering, chatbot automation, RAG systems, analytics dashboards
  • 98% client satisfaction with 40+ returning long-term clients
  • 500+ RLHF/SFT evaluation sessions on frontier models (Outlier, RWS, Translated)
  • Founded Wadware House in 2025 as a boutique AI automation agency

Stack

What I work with every week.

Not a tool inventory. These are the patterns and technologies behind the systems I ship.

AI & Agents

  • Multi-agent orchestration with human approval gates
  • OpenAI (GPT-4o, o1), Claude (Opus, Sonnet)
  • RAG pipelines: pgvector, Pinecone, LangChain
  • RLHF/SFT evaluation on frontier models
  • Prompt engineering and template management

Full Stack

  • React, Next.js, React Native (Expo)
  • Supabase: Postgres, Edge Functions, RLS, Realtime
  • TypeScript, Python, Deno, FastAPI
  • Stripe, Google Calendar, Gmail, Maps APIs
  • PostgreSQL, Redis, Firebase, pgvector

Infrastructure

  • GitHub Actions CI/CD, Docker, branch deploys
  • Sentry, Grafana, structured logging
  • n8n workflow automation, cron scheduling
  • Cloudflare, Vercel, Supabase hosting

Process

  • Architecture-first development with Claude Code
  • Team leadership: cross-functional, 3 to 8 people
  • Client-facing: requirements, SOWs, stakeholder comms
  • Built my own code review system (CodeLens)

Why this matters in 2026

Senior AI engineer demand is surging while entry-level roles flatten. Companies don't need more people who write code. They need people who design systems, manage AI agents, and make product decisions. That's the intersection where I operate: technical depth, system architecture, and the judgment to know what to build.

Roadmap

What I'm building next.

CodeLens v0.2

In Progress

Tree-sitter AST integration for semantic analysis, npm publish (npx codelens review), VS Code extension

AI Agent Orchestrator

Designing

Open-source framework for managing multi-agent workflows with planner/worker/validator pattern and human-in-the-loop gates

LLM Observability Dashboard

Planned

Self-hosted prompt tracking, token usage, latency monitoring, hallucination detection, and feedback loop analytics

Contact

Let's build something real.

Looking for an AI automation architect who ships production systems, not prototypes? I'd love to hear about your project.

Also available through Wadware House for agency engagements.