Job Openings
Fullstack AI Engineer (Onsite, Lahore, PKR Salary)
About the job Fullstack AI Engineer (Onsite, Lahore, PKR Salary)
Requirements:
- 4 years of experience as a fullstack or backend engineer
- Strong proficiency in Python and JavaScript/TypeScript
- Experience with FastAPI / Django / Node.js and React / Next.js
- Solid understanding of distributed systems and async architectures
- Hands-on experience deploying LLMs such as GPT-4/4.1, Claude, LLaMA, Mistral, Mixtral
- Experience serving models using vLLM, Triton, TGI, or similar frameworks
- Strong understanding of transformer models and inference trade-offs
- Experience with embeddings, vector search, and RAG architectures
- Experience with AWS, GCP, or Azure (GPU workloads preferred)
- Strong Docker and Kubernetes experience
- Familiarity with CI/CD pipelines for ML systems
- Experience with observability tools (Prometheus, Grafana, OpenTelemetry)
- Experience with multimodal AI (audio, video, image models)
- Experience optimizing LLM inference costs at scale
- Startup or high-growth environment experience
- Prior work on AI-first or AI-native products
Responsibilities:
- Deploy and optimize LLMs (open-source and commercial) for production use
- Implement inference optimization techniques (quantization, batching, caching, distillation)
- Build and maintain RAG pipelines (embeddings, vector databases, retrieval strategies)
- Evaluate and improve model quality (latency, accuracy, hallucination reduction, cost)
- Implement prompt management, versioning, and A/B testing
- Design and develop scalable APIs for AI-driven features
- Deploy and manage model-serving infrastructure (Docker, Kubernetes, GPUs)
- Optimize hardware utilization for inference workloads
- Implement monitoring, logging, and alerting for AI services
- Ensure security, data privacy, and compliance across AI pipelines
- Build internal tools and user-facing interfaces for AI workflows
- Integrate LLM services into web and mobile applications
- Work closely onsite with product managers, designers, and data teams
- Rapidly prototype, test, and iterate on AI-powered features