Full Stack Ai Engineer

 

Description:

This is a hands-on engineering seat. You'll work directly with the founder and own features from architecture decision to deploy to monitoring. The work spans multi-agent orchestration, retrieval and knowledge systems, document intelligence, voice AI, and the full-stack SaaS interfaces around them. Expect to ship across multiple products in the same week — and across all three clouds in the same quarter.

 

If you've shipped real AI to real users — and you're tired of POCs that never make it past the demo — this is the role.

 

What you'll do

  • Architect and ship multi-agent orchestration systems with supervisor routing, specialist agents, and tool use (Google ADK, AWS Bedrock AgentCore, LangGraph, CrewAI, or LangChain)
  • Build production RAG pipelines with hybrid retrieval, reranking, citation grounding, and anti-hallucination checks
  • Engineer document intelligence pipelines for extraction, classification, and validation across structured and unstructured inputs
  • Develop multi-tenant SaaS with zero-trust isolation, RLS, RBAC, audit trails, and atomic transaction integrity
  • Integrate LLM providers (OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI) with cost controls, latency budgets, and structured-output guardrails
  • Ship LLM features into existing tools — Office Add-ins, browser extensions, embedded iframes, real-time dashboards
  • Build voice AI integrations for autonomous calling and conversational agents (Twilio, Ultravox, LiveKit, Whisper, Deepgram)
  • Orchestrate complex workflows with n8n, Airflow, or Temporal
  • Deploy and operate on AWS, GCP, and Azure — Terraform, Bicep, or CDK for infra-as-code, GitHub Actions for CI/CD
  • Wire up observability — structured logging, distributed tracing, dashboards, alerts, runbooks
  • Make architecture calls and document them so the next engineer can ship from your code at 3 a.m.

 

What you bring

  • 2-4 years of professional engineering experience, with at least 1 year shipping production AI/LLM systems
  • Strong Python with FastAPI and async patterns at production scale
  • Frontend skill in Next.js 15+, React 18+, and TypeScript
  • Hands-on production experience with at least one agent framework — LangChain, LangGraph, Google ADK, CrewAI, or AWS Bedrock AgentCore
  • LLM API experience with at least two of: OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI
  • Vector database experience — Pinecone, Qdrant, Weaviate, Chroma, AstraDB, Milvus, or pgvector
  • Production deployment on at least one cloud (AWS, GCP, or Azure)
  • PostgreSQL or MongoDB at scale, Redis for caching, Docker, and CI/CD with GitHub Actions
  • Auth experience — JWT, OAuth, and at least one of Clerk, Firebase Auth, or Cognito
  • Bias for shipping. Boring, observable code. Production scars over framework demos.

Organization Cognilium AI
Industry IT / Telecom / Software Jobs
Occupational Category Full Stack AI Engineer
Job Location Lahore,Pakistan
Shift Type Morning
Job Type Full Time
Gender No Preference
Career Level Intermediate
Experience 2 Years
Posted at 2026-04-29 6:15 pm
Expires on 2026-06-13