Skip to content
Portrait of Tanzeel Naveed Khan

Tanzeel Naveed Khan

Forward Deployed Engineer (FDE)

Full Stack Developer · AI Engineer · Remote-first

About Me

Forward Deployed
Engineer.

FDE · Full Stack Developer & AI Engineer

I embed with your team, learn how the work actually gets done, and ship the system that does it, end to end, in production.

I'm Tanzeel Naveed Khan, a Forward Deployed Engineer (FDE) from Karachi, Pakistan, working as a full stack developer and AI engineer. FDE is a way of working more than a title. Instead of taking a spec over the wall, I sit inside the client's problem: their team, their data, their constraints. Then I build, deploy and iterate there until it holds up in production.

What I forward deploy is usually AI-native: Claude and OpenAI agents, skill-spec driven workflows and automation pipelines, built into an existing product rather than bolted on beside it. The engineering underneath is ordinary full stack work: Next.js, TypeScript, Python, Node. That is why it survives contact with real traffic.

The forward deployed part is what separates a demo from a system somebody depends on: deciding which tools an agent is allowed to call, keeping state durable outside the conversation, logging every call so a bad answer can be traced, and agreeing up front on what gets handed back to a human.

What I Bring as an FDE

A Forward Deployed Engineer covers the whole distance alone: understanding the workflow, architecting the system, writing the code, deploying it into your environment and staying on it while real users hit it. There is no handoff between the person who understood the problem and the person who built the answer.

In practice that means AI-native systems built with Claude and OpenAI agents, multi-agent orchestration and skill-spec driven development, standing on a full stack foundation engineered to be maintained rather than demoed.

How a Forward Deployed Engagement Runs

  1. Deploy in. I join your team, your tools and your codebase, and watch the workflow being done by the people who do it. The brief is almost never the problem.

  2. Ship a narrow slice. One real path, working end to end in your environment, early. A week of that teaches more than a month of specification.

  3. Widen and harden. Automate the boring path, leave the exceptions to a person, instrument everything, then hand it over documented, including why it is built the way it is.

50+

Projects Delivered

9+

Brands Worked With

3+

Years Experience

19+

Technologies

Frontend
Backend
Database
Deployment
Next.js
React
Python
Node.js
MongoDB
Stripe
Firebase
OpenAI
Claude AI logoClaude
Next.js
React
Python
Node.js
MongoDB
Stripe
Firebase
OpenAI
Claude AI logoClaude
Impact

How the Work Is Actually Built.

The engineering problem behind each forward deployed project, and the decision that kept it from becoming the usual mess.

The Challenge

The hard part of support automation is not producing an answer. It is deciding when not to answer, and when to hand the conversation to a human.

The Architecture

Classification, routing, and response generation are separate stages. A confidence threshold controls the routing decision, so low-confidence cases escalate automatically instead of shipping a wrong answer. Sentiment analysis feeds the routing input rather than sitting on top as a cosmetic feature.

Implementation & Docs

Prompts and classification rules live in configuration, so feedback from the support team applies without a deploy cycle. Every automated response writes a decision trail, which is what makes quality review possible after the fact.

Confidence Gated

Escalation

Config Driven

Prompt Layer

Cashbook AI

AI Systems

The Challenge

Automatic transaction categorization is worthless unless the user can correct a wrong call, and the system keeps that correction.

The Architecture

Categorization is a suggestion layer, not the final truth. The model's output is stored with its confidence, and the user override lives in a separate field, so the original prediction and the human correction are both preserved. Reports always derive from the final categorized state.

Implementation & Docs

Financial values are handled as integers so floating-point rounding errors never enter the ledger. The category taxonomy sits in configuration, letting a business map it to its own chart of accounts.

Correction Aware

Data Model

Precision Safe

Money Handling

The Challenge

Turn a voice message into text and return a context-aware LLM reply inside WhatsApp's webhook timeout, where every second counts against you.

The Architecture

The webhook receiver is separated from the processing pipeline, so WhatsApp gets an immediate acknowledgement while transcription and inference run in the background. The pipeline splits into three discrete stages: media fetch, speech to text, and response generation, each independently testable and replaceable.

Implementation & Docs

Every external dependency sits behind its own adapter module, so swapping a provider never touches business logic. The setup guide documents webhook verification, environment variables, and sandbox testing, so any developer can run the bot locally.

3 Stage

Processing Pipeline

Async

Webhook Handling

The Challenge

The real challenge in a no-code automation builder is not the UI, it is execution: running workflows safely, handling failures, and showing the user the result of every step.

The Architecture

A workflow is stored as a serializable graph, and the execution engine walks that graph node by node. Node types register through a registry pattern, so adding an integration never touches the engine. AI-assisted generation produces the same schema the manual builder does, rather than a second parallel system.

Implementation & Docs

Every execution writes step-level logs, which is where the dashboard derives both success rates and failure points. The node schema is documented so a contributor adding a node type is not guessing at the contract.

Registry Based

Node System

Step Level

Execution Logging

Experience

Brands I've
Worked.

Forward deployed with innovative companies across the globe, inside their teams, their codebases and their constraints.

Granule Services logo
Current Role

Granule Services

Full Stack Developer & AI Engineer

Leading AI powered applications built with Python and React.

2024 to PresentKarachi, Pakistan
AirGuard Envirocare logo
2026 to Present

AirGuard Envirocare

Full Stack Engineer

Building and maintaining full stack features end to end as a contract engineer.

2026 to PresentUnited Kingdom, Contract
Strombison logo
2026 to Present

Strombison

Full Stack Engineer

Building and maintaining full stack features across the product end to end.

2026 to PresentPakistan
DigitexHQ logo
2026 to Present

DigitexHQ

Full Stack Engineer

Building and maintaining full stack features across the product end to end.

2026 to PresentPakistan
Gyren logo
Past Role

Gyren

Backend Engineer

Building cutting edge technology solutions focused on innovation and scalability.

USA
Meetech Labs logo
Past Role

Meetech Labs

Backend & AI Engineer

Leading development of modern web, mobile, and software solutions for enterprise clients worldwide.

Mexico
DoorBeyond logo
Past Role

DoorBeyond

Full Stack Developer

Delivering modern digital experiences and e-commerce solutions that drive growth.

Pakistan
Seerah Tech logo
Past Role

Seerah Tech

App Developer

Providing specialized technology solutions and strategic consulting services.

UAE
Aviore logo
Past Role

Aviore

Backend Engineer

Focused on innovative technology solutions and digital transformation initiatives.

Pakistan
FAQ

Frequently Asked
Questions.

Everything people usually want to know before starting a project.

A Forward Deployed Engineer embeds inside the customer's team rather than working behind a ticket queue. They learn the real workflow first hand, then design, build, deploy and iterate the system in the customer's own environment until it holds up against real users and real data.

Yes. FDE is how the engagements here run: he joins your team, tools and codebase, watches the workflow being done, then ships the system end to end in your environment. It is the same person doing discovery, architecture, implementation and the iteration after launch.

Roughly three phases. Deploy in: time with the people doing the work, inside your repo and tooling. Ship a narrow slice: one real path working end to end, early. Widen and harden: automate the routine path, route exceptions to a person, instrument everything, hand over documented.

A contractor builds to a specification you wrote; an FDE helps you find out what the specification should have been, because they are in the room while the problem happens. There is also no handoff: the person who understood the problem is the person who writes and deploys the code.

Cost follows scope, so it is quoted per project rather than from a rate card. Most engagements run one of two ways: a fixed price against an agreed scope, or an ongoing contract when the work is continuous. Send the problem you are solving and you will get a scoped estimate back.

Both. Plenty of the work here is an addition to something already running rather than a rewrite, including custom booking logic built as a WordPress plugin so a theme update never touches it. The first step on existing code is reading it and telling you honestly what shape it is in.

Remote-first, with clients worldwide, from Karachi, Pakistan. Work is handed over in writing rather than depending on a shared working window, so progress does not stall waiting for an overlap. Project enquiries usually receive a reply within one working day.

Across the projects documented on this site: AI support and categorization systems, payroll and ERP platforms, healthcare booking, ecommerce and marketplaces, real-time chat, fleet and logistics tooling, ESG reporting, and manufacturer catalogues. The engineering write-up for each one is on the Impact page.

Tanzeel Naveed Khan is a Forward Deployed Engineer (FDE), full stack developer and AI engineer who builds AI-native web applications and intelligent automation systems embedded with the client's team. He works across the whole stack, from Next.js and React front ends to Python and Node.js back ends, and ships from concept through to deployment.

Forward deployed engineering (FDE) engagements, full stack web development, AI agent development and workflow automation. That covers embedding with a client team, building custom web applications, integrating AI models such as OpenAI and Claude into products, connecting payments and databases, and architecting systems designed to scale.

His core stack is Next.js, React, TypeScript and Tailwind CSS on the front end, with Node.js and Python on the back end. He also works with MongoDB, Firebase and Stripe, and builds AI features using the OpenAI and Claude agent APIs.

Yes. He is currently available for new projects and forward deployed engineer (FDE) roles, and works remote-first with clients globally. Engagements range from single-feature builds and AI integrations to embedding with a team and delivering a complete application end to end, on either a project or ongoing contract basis.

The fastest route is email at contact@tanzeelnaveedkhan.com, or the contact form on this site, which sends the message straight to his inbox. He is also reachable on WhatsApp and LinkedIn. Project enquiries usually receive a reply within one working day.

Get in Touch

Let's Build
Together.

Have a project in mind? Choose your preferred way to connect.

Send a Message

Fill out the form below and I'll get back to you soon.

Operating Globally · Remote-First · Available Worldwide