Pooya Golchian
Pooya Golchian is an AI Transformation Lead and AI Product Engineer based in Dubai. Builder of Hisabi.ai and Technova Solutions (codenovai.com). Shipping software since 2009, with the last three years in production AI. Master of Science in Software Engineering. I take B2B teams from AI strategy to production, designing agent systems, deploying private LLMs, and shipping full-stack Next.js and Node.js platforms that hold up under load.
The work, in order.
My first paid code shipped in 2009, freelance work that put me directly into client deadlines while I was still studying. I went on to earn a Master of Science in Software Engineering, which sharpened the discipline behind every system I have built since. The first stretch of full-time product work was frontend architecture across Angular, then Vue, then React, building admin panels and customer-facing apps for an e-commerce, a content company, and the InsurTech market in Iran. Different stacks, same job. Turn a designer's intent into something that loads fast on the worst connection in the room.
The middle stretch hardened the discipline. Five years across Azki and Azkivam, two regulated FinTech platforms, rising from senior engineer to leading a fifteen-person pod. A real-time financial dashboard on Vue and Electron streaming market data over WebSockets at sub-second latency. A 120-component Storybook library in React and TypeScript that lifted delivery speed across the organisation. SSR work that took Lighthouse from 55 to 95 in a quarter.
Cashew Payments in Dubai was where the senior delivery muscle locked in. Payment gateway integrations with 99.99% uptime, fraud reduction work tied to the SEON SDK, and an Nx monorepo with seventy-plus reusable components that cut team delivery time by forty percent. Payment systems handling $10M+ per month across thousands of daily users. I ran accessibility and localization to WCAG 2.1 because regulated markets don't ship without it.
Three years ago the AI work began in earnest. Not as a side topic, as the centre of gravity. I architect production agent systems with LangGraph and the Vercel AI SDK, design Model Context Protocol servers that expose internal data to AI clients safely, and build RAG pipelines that survive document churn. I write eval suites before shipping. I worry about token economics the way I used to worry about CDN costs. I deploy private LLMs on customer infrastructure for industries where data residency is the only acceptable answer.
The work stays full-stack. Most engagements pair AI engineering with the rest of the stack the team is running: Next.js App Router, Node.js services, monorepos, durable workflows, the integration layer that makes a feature feel like one product instead of three bolted together. I write code my colleagues can own after I leave.
I run two companies in parallel with the consulting practice. Hisabi.ai builds AI products for B2B teams shipping into regulated markets, and Technova Solutions (codenovai.com) is the engineering studio behind several of the production deployments listed in my case studies. I own delivery at both, which is why I lead AI transformation rather than just write code. The job is to move a team from AI strategy to a shipped product they can own, not just the code path.
Eight companies. Industries from fintech to insurtech to luxury retail to live gambling to AI-native B2B SaaS. The thread is the same line of work: senior engineering, shipped, in production, under deadline.
The skills I'm currently building against.
The market has moved fast since late 2024. These are the surfaces real B2B teams are paying premium rates for right now. They're also the surfaces I work in daily, on real customer engagements.
Agentic systems
- LangGraph & multi-agent orchestration
- OpenAI tool calling + structured outputs
- Vercel AI SDK with streaming UX
- Anthropic Claude with extended thinking
MCP & integration
- Custom MCP server design
- Auth, rate limits, and capability scoping
- Tool surface testing & versioning
- Claude Desktop and IDE clients
Evals & observability
- Golden datasets + LLM-as-judge calibration
- Braintrust, LangSmith, Helicone tracing
- A/B harness for prompts and models
- Drift detection on production traces
RAG & retrieval
- Hybrid search with reranking
- Citation-first response design
- Chunking strategies that survive churn
- pgvector, Weaviate, Pinecone
Inference & cost
- vLLM, Ollama, on-prem serving
- Token economics & model routing
- Quantization, LoRA, QLoRA fine-tuning
- Open-weight models (Llama, Qwen, DeepSeek)
Voice & multimodal
- Realtime API & low-latency voice agents
- Vision pipelines for document AI
- Whisper, Eleven Labs, Deepgram
- Image generation (Flux, Midjourney v7)
Full-stack delivery
- Next.js 16 App Router + Server Components
- Node.js services & durable workflows
- Nx + pnpm monorepos
- AWS via SST and OpenNext
AI safety & guardrails
- Prompt-injection defenses
- PII redaction & per-role access
- Content moderation pipelines
- Audit logs for regulated sectors
Where the time went.
Reverse chronological. Reading bottom to top is the actual order things happened.
AI Transformation Lead & Forward Deployed Engineer
Cashew Payments & HazelPay (FinTech / BNPL / LendTech)
- Built the BNPL business end to end from scratch, then rose to lead its AI transformation. Wrote the backend in Java and Spring Boot across 18 microservices on PostgreSQL, added Python and FastAPI services for the scoring and inference workloads and Go for the high-throughput paths, and built the customer and admin frontends in Next.js and React.
- Made the call to leave the microservice estate behind and led the migration myself, applying the Strangler Fig pattern to move traffic service by service into a Turborepo monorepo with a consolidated backend and a Backend-for-Frontend layer. New infrastructure, 35% less duplicated code, and a codebase shaped so agents could work in it.
- Designed the AI-Driven Development Lifecycle that estate now runs on. Machine-readable specs drive Claude Code agents through generation, review and test behind evaluation harnesses and approval gates. Founded the practice, trained 25 engineers on subagent design and MCP tooling, and cut cross-team delivery time by roughly 40%.
- Built the fraud, KYC and risk stack running across ten products under the Mashreq Bank partnership. Trained and deployed supervised models in Python over engineered feature pipelines for risk scoring and transaction monitoring, then combined them with vendor signals from Shufti, Emirates Face Recognition and SEON behind an agent fleet wired through MCP servers. Every automated decision clears a human review queue and writes an explainable trace, and fraudulent account creation fell 30%.
- Carried the operating model past engineering, which meant sitting with each function rather than shipping a tool. Interviewed and onboarded 15 people across HR, Finance, Product Design, Marketing and Sales onto Claude Code with skills built for their own daily work, backed by a shared knowledge base so context travels between teams.
- Built HazelPay on AWS with SST v3 and Next.js, unifying admin, dashboard and customer portals onto one platform, and Pinapl (pinapl.ai), an AI marketing agent for Shopify merchants that runs abandoned-cart and reactivation campaigns autonomously under human approval.
JavaSpring BootPostgreSQLNext.jsReactTurborepoClaude CodeMCPAWSForward Deployed Engineer
Independent AI Consulting (Technova Solutions FZCO)
- Embedded with UAE enterprise clients as their lead engineer, working on their infrastructure and inside their compliance constraints. Stood up NVIDIA GPU inference on customer premises and fine-tuned Arabic models, Jais and Falcon, for organisations whose data cannot legally leave the country under PDPL.
- Built on-premise Retrieval-Augmented Generation pipelines in Python answering in under three seconds, with citation-first responses so a client reviewer can check any claim against its source. Handed each deployment over to the customer's own team.
- Built the batch and streaming data pipelines these systems run on, moving client data through PySpark, pandas and Polars into BigQuery and PostgreSQL. Tuned partitioning and clustering so analytical queries stay affordable at enterprise volume, and ran the surrounding services on GCP alongside AWS.
- Designed and delivered Marteh, an agentic sales platform for GCC real estate, running voice agents in Arabic, English, Hindi and Urdu against live property data. Deal-cycle time fell 40% for the client.
- Delivered Hisabi, a bilingual invoicing platform with voice-to-invoice, OCR and UAE VAT compliance, on AWS ECS Fargate provisioned through Terraform.
LangChainLangGraphRAGTerraformECS FargatePrismaPythonSenior Software Engineer & Engineering Pod Lead
Azki (InsurTech) / Azkivam (LendTech)
- Rose from senior engineer to leading a 15-person pod across two regulated FinTech platforms, owning technical direction, delivery standards and mentoring.
- Shipped a component library of 120+ React and TypeScript pieces in Storybook, which lifted feature-delivery speed across the organisation by 40%.
- Rebuilt the Next.js and Nuxt.js server-rendering pipelines on high-traffic insurance and price-comparison sites, taking Lighthouse performance from 55 to above 95.
- Architected a real-time financial dashboard in Vue and Electron, streaming market data over WebSockets at sub-second latency under heavy concurrent load.
Node.jsReactVueNuxt.jsStorybookElectronWebSocketFull-Stack Developer (C# & PHP)
Axprint (E-commerce)
- Built the company's first React storefront and admin order-management panel, setting a frontend architecture that carried the business for five years.
- Delivered catalogue, checkout and order-management flows across a C# (.NET) and PHP (Symfony) backend for a high-traffic store.
ReactC# (.NET)PHP (Symfony)Angular 1jQuery
How I work.
Engineer first, consultant second. Most engagements run as fractional senior engineering: two to three days a week for four to twelve weeks, embedded with the team rather than over the wall. The deliverable is shipped code your engineers can own after I leave, plus the documentation and evals that prove it works.
No no-code tools. No vibe coding. No vendor lock-in baked into the architecture because it was easier in week one. The market has cooled on agencies that ship demos in pretty wrappers, and rightly so. Production AI today is engineering work, which is what the next twelve years will keep being.
Questions people ask about Pooya.
- Who is Pooya Golchian?
- Pooya Golchian is an AI Transformation Lead and AI Product Engineer based in Dubai, UAE. He takes B2B teams from AI strategy to production and has been shipping software since 2009, with the last three years focused on production AI. Pooya holds a Master of Science in Software Engineering and is the founder of Hisabi.ai and Technova Solutions (codenovai.com).
- What does AI transformation with Pooya Golchian involve?
- Pooya Golchian leads AI transformation as hands-on engineering, not slideware. He works with B2B teams to pick the workflows worth automating, build the agent systems and RAG pipelines that run them, deploy private LLMs where data residency matters, and train the in-house team to own the result. Engagements run fractionally, two to three days per week, embedded with the team.
- What does Pooya Golchian do?
- Pooya builds production AI systems for B2B teams. His work covers agent systems with LangGraph and the Vercel AI SDK, Model Context Protocol (MCP) servers, RAG pipelines with citation-first responses, evals with golden datasets and LLM-as-judge, private LLMs deployed on customer infrastructure, and full-stack Next.js platforms on AWS via SST and OpenNext.
- Where is Pooya Golchian based?
- Pooya Golchian lives and works in Dubai, UAE. He takes on remote and embedded engagements across fintech, insurtech, luxury retail, live gambling, and AI-native B2B SaaS.
- What is Hisabi.ai?
- Hisabi.ai is the company Pooya Golchian founded to ship AI products for B2B teams in regulated markets. It focuses on production-grade agent systems, RAG with citations, and private LLM deployments where data sensitivity and audit trails matter.
- Who built Cashew Payments?
- Pooya Golchian built the platform behind Cashew Payments during his fintech work. The platform processes more than $10M per month in payment volume at 99.99% uptime.
- How does Pooya Golchian work with clients?
- Pooya works fractionally, two to three days per week for four to twelve weeks, embedded directly with the client team. The model suits B2B teams that need senior AI product engineering without a full-time hire. Engagement details and contact form live at /contact and /services.
Available for fractional engagements.
Two to three days a week, embedded for a quarter. Tell me what you're shipping.