Who is Fahad Murtaza?
Fahad Murtaza is a software engineer with 23 years of full-stack experience, currently specializing in AI evaluation engineering and WordPress AI integration. As an expert task author in Mercor's SWE-bench-Extended program, he designs Docker-reproducible benchmark tasks that frontier AI labs use to grade code agents across eight programming languages. He is the founder of iSuperCoder, a software development firm, and a Top 3% developer on Toptal. His open-source tools include elementor-mcp — a 70-tool Model Context Protocol server connecting Elementor to AI coding agents such as Claude Code and Cursor AI — and WordPress Hooks Explorer, which indexes 3,254 WordPress core hooks into a searchable reference. He has shipped production software for Propertyfinder.ae, BackTable, and 100+ SMB clients across logistics, real estate, healthcare, and tourism. He works remotely from the United Arab Emirates and takes on a limited number of AI evaluation and WordPress consulting engagements each quarter.
Proof & Track Record
- 130+ SWE-bench-Extended tasks authored across 8 languages, work consumed by frontier AI labs (Mercor expert program, Q1 2026)
- ≥0.95 QC threshold sustained across all task submissions, both reviewer and super-reviewer scores
- 100+ shipped client projects across logistics, real estate, tourism, events, healthcare, lending
- Distinctive AI tooling:
elementor-mcp(70-tool MCP server) andwordpress-hooks-explorer(3,000+ WP hooks indexed) - Free educational platform: Rust 90 Days, Algorithms in 60 Days, Claude Code for Beginners (15 modules)
What services does Fahad Murtaza offer?
AI Evaluation & Rubric Engineering
Agentic-benchmark task authoring, LLM-as-judge pipeline design, eval-harness consulting for AI labs and the data-labeling supply chain.
AI Integration Audits
An AI integration audit is an independent technical review of GPT or Claude API calls already running in production code. The review covers four dimensions: prompt quality (clarity, injection risk, hallucination surface), fallback architecture (what happens when the API times out or returns a refusal), cost ceilings (estimated spend per request at current call volume, opportunities for prompt caching or batching), and failure-mode visibility (whether errors surface in observability tooling or fail silently to users). Output is a written report with severity-ranked findings and concrete recommendations — typically model selection, streaming vs. batch, caching layer design, or RAG vs. fine-tuning trade-offs. Engagements are fixed-bid at one week of review time. Common triggers: a production integration behaving inconsistently, LLM costs exceeding projections, or a team inheriting AI code they didn't author. Most audits complete within five business days of receiving code access.
WordPress + AI Feature Builds
WordPress AI feature builds integrate large language models directly into existing WP sites without requiring a full platform migration. Common deliverables include: chatbots embedded in page templates that answer questions from a site's own documentation via retrieval-augmented generation (RAG); content generation tools wired into the WP admin that draft posts, product descriptions, or emails using Claude or GPT-4; and MCP servers that expose WP admin capabilities to AI coding agents. The elementor-mcp project, for example, wraps 70 Elementor admin actions into a standardized MCP server that Claude Code and Cursor AI can call directly. The wc-ai-chatbot routes WooCommerce shopping queries across Claude and Moonshot via dual-LLM logic, falling back automatically when one provider is unavailable. Engagements are scoped as fixed-bid 2–6 week builds once requirements are clear. Every build ships with documentation and a handover call so in-house teams can maintain it.
WordPress Integrations
WP ↔ Stripe, PayPal, GiveWP, MemberPress, Brevo, MLS feeds, custom CRMs. Fixed-bid 2-4 week engagements.
Specialist Plugin Work
Gravity Forms, WooCommerce, MEC, Eventin, LifterLMS extensions. 6+ shipped Gravity Forms plugins alone.
WP Hosting & Dev Environments
Docker stacks, LEMP automation, server migrations, performance optimization with Caddy / Nginx.
Currently Shipping (2026)
Mercor Agentic-Bench
Expert task author for SWE-bench-Extended. 130+ tasks across 8 languages. Customers under NDA.
Visit Grand Rapids
Custom WP Travel theme + plugins for the city's tourism site. Demo hosting via wp-launcher.
AJS Trucking
Multi-plugin engagement: toolkit, driver dropdown, date migrator, date fix, GF integrations.
iSuperCoder Platform
Rust microservices (auth + user, 494 tests) behind a Next.js 15 gateway. isupercoder.com
elementor-mcp
70-tool MCP server bridging Elementor to AI clients (Claude Code, Cursor, etc.).
wc-ai-chatbot
WooCommerce AI shopping assistant, Claude + Moonshot dual-LLM routing.
Technology
Languages I ship in
Frontend frameworks
Backend frameworks
AI / LLM
Polyglot reading depth
WordPress
Infra & DevOps
Frequently Asked Questions
What is SWE-bench-Extended task authoring?
SWE-bench-Extended task authoring is the process of converting real GitHub issues or pull requests from open-source projects into structured evaluation tasks for AI coding agents. Each task includes a pinned Docker environment, structured problem and prompt statements, an interface contract, a golden patch, a test patch, and an implementation-agnostic rubric covering functional, robustness, and style criteria. I have shipped 130+ such tasks across Go, Rust, Java, Kotlin, C++, JavaScript, TypeScript, and Python through Mercor's expert program in Q1 2026, sustaining a ≥0.95 QC threshold.
What does an AI integration audit cover?
An AI integration audit is a fixed-bid one-week review of GPT or Claude API integrations already running in production code. It covers four areas: prompt quality and injection risk, fallback architecture, per-request cost analysis, and failure-mode visibility in observability tooling. Output is a written report with severity-ranked findings and concrete recommendations including model selection, prompt caching, streaming, or RAG vs. fine-tuning trade-offs.
What WordPress AI features can Fahad Murtaza build?
I build WordPress AI features including site-specific chatbots using retrieval-augmented generation (RAG), content generation tools wired into the WP admin, WooCommerce AI shopping assistants, and Model Context Protocol (MCP) servers that expose WP admin capabilities to AI coding agents. My elementor-mcp project provides a 70-tool MCP server bridging Elementor to Claude Code and Cursor AI.
How do I hire Fahad Murtaza?
I am available for AI evaluation contracts, integration audits, and WordPress AI builds. Engagements are fixed-bid where scope is clear and hourly where it is not. Reach me at [email protected], via Toptal, or by scheduling a call. I respond to qualified inbound within 24 hours.
What has Fahad Murtaza shipped in 2026?
Working with Me
AI eval contracts, integration audits, and WordPress + AI builds. Fixed-bid where scope is clear; hourly where it isn't.
Start a Conversation ↗Get an Agentic Evaluation →Speaking & Training
International speaker at Google, Python conferences, and WordCamps on practical AI in software development. Trains engineering teams on Flutter and AI coding tools including Claude Code, Cursor AI, and Gemini.