Job Description & Role Mission
Arbisoft’s Applied Artificial Intelligence Laboratory is looking for an extraordinary Lead AI & Machine Learning Engineer to spearhead generative AI engineering, custom fine-tuning of open-weights language models, and agentic workflows for enterprise customers in Silicon Valley, edX, and healthcare sectors.
In this position, you will design state-of-the-art Retrieval-Augmented Generation (RAG) pipelines, implement intelligent reasoning agents, optimize vector embedding pipelines, and quantize large neural models for ultra-low latency real-time inference on cloud and edge GPU clusters.
You will work with high autonomy alongside a team of elite computer scientists, research fellows, and product innovators. We value foundational problem-solving, intellectual rigor, reproducible empirical research, and code that is clean, modular, and built for immense real-world scale.
Key Responsibilities
- Design, build, and deploy production-grade RAG architectures using LangChain, LlamaIndex, vLLM, and hybrid vector/lexical retrieval.
- Fine-tune open-source foundational models (Llama 3, Mistral, Gemma) using LoRA, QLoRA, and parameter-efficient techniques on NVIDIA H100/A100 clusters.
- Develop multi-agent collaboration frameworks (AutoGen, CrewAI) capable of autonomous tool-use, code execution, and complex reasoning.
- Implement robust evaluation frameworks (RAGAS, TruLens) for measuring model hallucinations, faithfulness, and semantic retrieval accuracy.
- Benchmark and optimize model inference latency via TensorRT-LLM, ONNX Runtime, and KV-cache optimizations.
- Collaborate with backend engineers to integrate AI models seamlessly into microservices architectures via FastAPI and gRPC.
- Keep abreast of cutting-edge AI breakthroughs in multi-modal vision-language models, reasoning models, and agent architectures.
Requirements & Qualifications
- BS, MS, or PhD in Computer Science, Artificial Intelligence, Data Science, or related quantitative field.
- 3+ years of production experience in machine learning, NLP, and deep learning architectures.
- Proficiency with Python, PyTorch, Hugging Face transformers, and vector databases (Qdrant, Milvus, Pinecone, or pgvector).
- Demonstrated experience building end-to-end RAG systems with rerankers, contextual compression, and chunking strategies.
- Experience with distributed training, GPU memory profiling, and quantization (AWQ, GPTQ, GGUF).
- Strong software engineering fundamentals: modular design, testing, Docker, and CI/CD pipelines.
- Published research papers in top-tier conferences (NeurIPS, ICML, CVPR, ACL) or strong GitHub open-source contributions is a major plus.
Required Skills & Competencies
PythonPyTorchLarge Language Models (LLMs)RAGLangChainvLLMHugging FaceVector DatabasesFastAPIDocker
Benefits & Compensation
Top-tier market compensation with bi-annual performance and salary evaluations
Comprehensive Health Insurance covering employee, spouse, children, and parents with 100% OPD reimbursement
Free freshly prepared gourmet lunch, snacks, and specialty coffee on campus
Company annual international retreat and family adventure trips
Provident Fund with matching contributions and zero lock-in penalty
Flexible working hours with generous work-from-home allowances
Unlimited books purchase allowance and full sponsorship for technical conferences
About Arbisoft
Founded in 2007, Arbisoft is one of Pakistan’s most prestigious technology companies, building world-class software for global tech leaders including edX, Stanford University, KAYAK, and enterprise startups. Arbisoft is renowned for its people-centric engineering culture, academic partnerships, and pioneering research in artificial intelligence and scalable distributed systems.
Direct On-Site Application
Ready to apply for Lead AI & Machine Learning Engineer (LLMs, RAG & Agents)?
Submit your resume and contact information directly through Pakistan Digital. Your application is securely transferred to the hiring team at Arbisoft without being redirected.
⚡ Fast 1-Minute Apply🔒 Free & 100% Confidential