Full-Stack JavaScript Engineer — AI Agents & Automation

Full-Stack JavaScript Engineer — AI Agents & Automation

Location: Gulshan-e-Iqbal, Karachi — On-site
Timing: 8:00 PM–5:00 AM
Salary: PKR 100,000/month – PKR 125,000/month
Company: Outsource Origin

We are looking for a Full-Stack JavaScript Engineer to join our product team and build production SaaS platforms using Node.js, TypeScript, React/Next.js, LLMs, autonomous AI workflows, tool calling, and external API integrations.

IMPORTANT — READ BEFORE APPLYING

This position requires real development experience with the technologies mentioned below.

Please do not apply if your AI development experience comes primarily from tutorials, university projects, courses, hackathons, simple chatbot demos, or experimental projects.

We need someone who has actually built, debugged, deployed, and maintained real applications.

Required Technical Experience

• Strong Node.js / TypeScript
• Strong React.js / Next.js
• Backend API development and database architecture
• LLM-powered application development
• AI agents and multi-step AI workflows
• LLM function/tool calling
• RAG, embeddings, and vector databases
• Routing requests between different tools, APIs, or agents
• Maintaining state across multi-step workflows
• Background processing and job queues
• Redis / PostgreSQL or equivalent technologies
• REST APIs, webhooks, and external service integrations
• Docker and production deployments
• Retry logic, error handling, monitoring, and failure recovery

Hands-on experience with LangGraph, OpenAI Agents SDK, MCP, BullMQ, Temporal, Celery, or similar frameworks/tools will be considered a strong advantage.

What You Will Build

We're developing SaaS products where AI is part of the application's actual workflow rather than simply a chatbot added to the interface.

For example, one request might require the system to:

Analyze Request → Determine Required Actions → Choose Tools/Agents → Retrieve Data → Execute Multiple Tasks → Call APIs → Generate Output → Store State → Recover From Errors → Complete the Request

These processes may involve several models, APIs, database operations, background workers, and external services.

Some tasks finish within seconds. Others may continue processing for several minutes.

You need to understand how to build these workflows reliably, including what happens when an API times out, an AI step produces an invalid response, a worker crashes, or one step fails halfway through execution.

Who We're Looking For

We want an engineer who understands what is happening behind the interface, not someone who only knows how to connect an LLM API to a frontend.

You should be able to clearly explain systems you personally built, including:

• Application architecture
• Backend design
• Agent/tool routing
• Function calling
• Database and state management
• Queues and workers
• Retry/failure handling
• Third-party API integrations
• Deployment and production debugging

Our development team is intentionally small, so engineers are expected to take significant ownership.

That means you may take a feature from initial architecture and implementation all the way through testing, deployment, debugging, and production improvements.

If you enjoy building complete systems instead of working on one small isolated component, this role may suit you.

Pay: Rs100,000.00 - Rs125,000.00 per month

Application Question(s):

• Do you have professional experience building applications using both Node.js/TypeScript and React/Next.js?
• Have you personally developed a real-world AI application using tool/function calling, RAG, AI agents, or multi-step LLM workflows? Please exclude coursework, tutorials, hackathons, and personal demos.
• Describe the most technically advanced AI system you have personally built. What happened after a user submitted a request, how were different tools/agents selected, and which parts did you personally develop?
• Have you implemented production systems involving background workers, queues, persistent workflow state, retries, failure recovery, or long-running jobs? If yes, briefly explain how your architecture worked.
• Outsource Origin started in 2022 as a service company and has recently moved toward building its own software products. Our product team is currently small, and engineers are expected to own problems across architecture, coding, debugging, and deployment. Are you comfortable working in this type of environment?
• This is a full-time, on-site role in Gulshan-e-Iqbal, Karachi, working 6 days per week from 8:00 PM–5:00 AM. The starting salary is PKR 100,000/month. Are you comfortable with the location, schedule, shift timing, and salary?
• If selected, are you able to join immediately or within 1–2 days? During the interview, can you also demonstrate real applications you personally worked on and explain permitted code/architecture from those projects?

Work Location: In person