Senior engineering for the systemsyour business depends on.
FlintLark is a Dubai-based engineering studio shipping AI platforms, custom software, mobile apps, and cloud infrastructure — for funded startups, regulated enterprises, and ambitious SMBs across the UAE, UK, US, and Europe.
- Dubai · London
- 10+ years engineering
- $100M+ in production
- 15+ AI models live
- Insurance · Payments · Telecom · SaaS
From AI agents to full-stack platforms — we ship the lot.
Eight engineering capabilities under one roof. Mix and match to build the system your business needs — with AI baked in where it matters.
AI Agents & LLM Platforms
Multi-agent systems, RAG, fine-tuning, prompt engineering
Custom Software & Web Apps
Next.js, React, FastAPI, Spring Boot, NestJS
Mobile Applications
Native iOS/Android, React Native, cross-platform
Cloud & DevOps
AWS, GCP, Azure, Kubernetes, Terraform, IaC
Data & RAG Pipelines
Vector DBs, ETL, analytics, data engineering
Automation & Integrations
n8n, APIs, MCP, custom workflow engines
AI Compliance & Safety
Guardrails, EU AI Act, audit trails, governance
Technical Leadership
Fractional CTO, architecture audits, AI strategy
Battle-tested engineering, AI-fluent delivery.
FlintLark FZCO is the engineering studio of Petar Barzev — a Principal AI & Full-Stack Engineer with over a decade of experience building production systems at scale. Behind FlintLark sits a flexible network of senior engineers and trusted specialists. We scale up and down to match the project — never outsourced shop, always embedded partners.
From multi-agent AI platforms and LLM-powered automation to mobile apps, cloud infrastructure, and bespoke web platforms — we ship production-grade systems for clients across insurance, payments, telecom, SaaS, crypto, and consumer commerce. Based in Dubai with operations spanning the UK, US, and Europe, we take on work that other teams won't or can't.
What We Build
End-to-end engineering for companies that need to ship serious software — with AI built in where it matters, and without it where it doesn't.
AI Agents & LLM Engineering
Production-grade AI: autonomous agents, multi-agent orchestration, RAG architectures, LLM fine-tuning, and prompt engineering across Claude, GPT-5, Gemini 3, Llama 4, DeepSeek, and Mistral. We build the AI layer — and the engineering rigor that makes it stay up.
Custom Software & Web Apps
End-to-end product builds with Next.js, React, FastAPI, Spring Boot, NestJS, and cloud-native infrastructure. Pixel-perfect frontends, bulletproof APIs, and the architectural decisions that don't blow up in year two.
Mobile & Cross-Platform Apps
Native iOS and Android, React Native, and Flutter. From MVP to App Store launch — including push, payments, offline sync, and the AI capabilities that make modern apps actually useful.
Cloud Architecture & DevOps
Platform engineering on AWS, GCP, and Azure. Kubernetes, Terraform, CI/CD pipelines, observability, internal developer platforms, and FinOps. We build the infrastructure your engineering team will thank you for.
Intelligent Automation & Integrations
Replace manual processes with autonomous systems. n8n, custom Python pipelines, MCP-based tool integration, and AI-native workflow engines — from email triage and document processing to full end-to-end business process automation.
Technical Leadership & AI Strategy
Fractional CTO, architecture audits, AI strategy, team scaling, and 2–4 week paid discovery sprints for enterprise buyers. The senior engineering perspective your roadmap needs — without the full-time hire.
DealHabibi.com Has Launched
The UAE's first community-driven deals platform is live. Join thousands of shoppers discovering the best deals across Amazon.ae, Noon, Carrefour, and 500+ retailers — powered by real community votes and verified buying experiences.
Built and shipped in 60 days. The latest example of what FlintLark does for clients every week.
Things We've Shipped
Real products. Real users. Real production traffic. Real results.
Skylar
AI Sales Coach used by Databricks-tier B2B teams
1,500+ practice calls/week · 30+ languages
Co-architected and built the core AI engine for Skylar — an intelligent sales coaching platform used by B2B teams at companies like Databricks and Wove. Designed the LLM orchestration layer powering AI role-play personas trained on real buyer interviews, built the RAG pipeline ingesting company-specific sales playbooks and objection libraries, and engineered the real-time NLP feedback system that analyzes rep performance across 30+ languages.
Payment Interfaces
Powering Netflix · ASOS · Uber Eats · 50,000+ merchants
$100M+ processed · PCI DSS · Globally deployed
Core contributor to Checkout.com's Payment Links, Hosted Payments Pages, and Frames SDK — the payment interface layer used by Netflix, ASOS, Uber Eats, and thousands of global merchants. Built accessible, PCI-compliant checkout experiences at massive scale, with ML-driven fraud scoring and conversion optimization.
DealHabibi
The UAE's Community-Driven Deals Platform
500+ retailers · Live for Ramadan 2026
A community platform where UAE shoppers discover deals, share real buying experiences, compare prices, and vote on the best offers across 500+ retailers. Features community voting, affiliate integration with Amazon.ae and Noon, real-time deal submission with URL scraping, content moderation for UAE compliance, and privacy-first analytics — built from scratch with Next.js, NestJS, PostgreSQL, and Cloudflare.
MeowGTP
Multi-Model AI Chat Platform
15+ LLMs · Thousands of conversations daily
A unified AI chat interface integrating 15+ large language models including GPT-5, Claude Opus, Gemini 3, Mistral, Llama 4, and DeepSeek. Features real-time streaming responses, intelligent model routing, usage analytics, and PurrSafe — a family-friendly AI content guardrail system powered by multi-layer LLM classification. Serving thousands of conversations daily.
Nivrin
AI-Powered Creative Platform
Multi-model generation pipeline
An intelligent content creation platform that transforms images into professional poster designs using curated AI prompt chains, custom templates, and an optimized multi-model generation pipeline — combining AI automation with human quality assurance for flawless results.
Technologies We Work With
AI / ML
- OpenAI GPT-5
- Anthropic Claude
- Google Gemini
- Meta Llama
- DeepSeek
- Mistral
- Hugging Face
- PyTorch
- scikit-learn
AI Agents & Orchestration
- LangChain
- LangGraph
- CrewAI
- AutoGen
- MCP Protocol
- LlamaIndex
- Haystack
AI Infrastructure
- RAG
- Pinecone
- Weaviate
- Qdrant
- pgvector
- NeMo Guardrails
- Guardrails AI
- MLflow
- Weights & Biases
Automation
- n8n
- Custom Pipelines
- Zapier
- Make
Frontend
- Next.js
- React
- Angular
- Vue.js
- TypeScript
- Tailwind CSS
- React Native
Backend
- FastAPI
- Django
- Flask
- Spring Boot
- Quarkus
- NestJS
- Node.js
- Express
- gRPC
- GraphQL
- Prisma
Languages
- TypeScript
- Python
- Java
- Kotlin
- Swift
- JavaScript
Cloud & DevOps
- AWS
- GCP
- Azure
- Docker
- Kubernetes
- Terraform
- Vercel
- Cloudflare
Data
- PostgreSQL
- MongoDB
- Redis
- Supabase
- Elasticsearch
Payments
- Stripe
- Checkout.com
- PCI DSS
Where We've Made Impact
Engineering that has shipped at scale
FlintLark FZCO
·Founder & Principal EngineerEngineering studio shipping AI platforms, custom software, mobile apps, and cloud infrastructure. Production-grade AI agents, RAG pipelines, AI sales coaching engines, and content generation systems. Embedded engineering partner for clients across the UK, US, Europe, UAE, and beyond.
Skylar
·Co-Architect & Lead AI EngineerCo-architecting the AI engine behind Skylar — an AI sales coaching platform used by B2B teams globally. Built the LLM orchestration layer for AI role-play personas, RAG pipeline for sales playbook ingestion, NLP-based conversation analysis for real-time feedback, and multi-language AI support across 30+ languages.
Leading Lloyd's of London Syndicate
·Senior Engineer (AI & Full-Stack)Architecture and delivery across an algorithmically underwritten broker platform at global scale. Integrating LLM-powered automation flows for underwriting decision support, building RAG pipelines for policy document analysis, and engineering NLP-based broker submission parsing — reducing manual analyst workloads and improving operational turnaround.
BT Group
·Platform EngineerPlatform engineering for one of the world's leading telecommunications companies. Built internal developer portals with Port.io, Kubernetes-based service templates and flavours on AWS, CI/CD pipelines, monitoring dashboards with Dynatrace, PagerDuty alerting integrations, and self-service infrastructure tooling enabling engineering teams to launch production services in minutes.
Checkout.com
·Senior Software EngineerContributed to Payment Links, Hosted Payments Pages, and the Frames SDK — serving Netflix, ASOS, Uber Eats, and 50,000+ merchants globally. Built accessible, PCI-compliant checkout experiences processing hundreds of millions in volume. Drove ML-powered fraud detection tuning and conversion optimization, contributing to higher acceptance rates.
Uphold
·Senior Backend EngineerBackend engineering for a multi-asset digital money platform supporting crypto, fiat, and precious metals trading across 150+ countries. Built high-throughput transaction services with gRPC microservice patterns and contributed to Kubernetes orchestration at scale.
Let's build something remarkable.
Tell us about your project — AI platform, custom software, mobile app, cloud architecture, or something else entirely. We respond within one business day.
Book a 30-min intro callTypical reply: within 1 business day · EMEA business hours