FIND_THE_RIGHTJOB.
Senior AI Engineer
JOB_REQUIREMENTS
Hires in
Not specified
Employment Type
Not specified
Company Location
Not specified
Salary
Not specified
Last Date: Sunday, January 4, 2026
Job Detail
Job Ref #:
Job/5986/12/23/2025
Age Limit:
2545
Experience:
Posted Date:
Tuesday, December 23, 2025
Salary:
Market Competitive
Job Description:
Qualification:
- BS degree in Computer Science, Software Engineering, or Computer Engineering from an HEC recognized University.
Experience:
- 3 to 4 years of hands-on experience building and shipping AI systems in production.
- Hands-on experience training and fine-tuning LLMs, including LoRA-based fine-tuning.
- Proven experience building RAG pipelines (chunking, embeddings, retrieval, reranking, grounding, and evaluation,
- Experience building agentic systems using LangChain and LangGraph, deploying models locally on GPUs using vLLM and/or Ollama, Ray and Kubernetes for scalable serving & operations and building real-time voice agents using LiveKit.
- Experience with automated evaluation and benchmarking for LLM, RAG, and agent workflows.
- Experience implementing guardrails and secure tool execution patterns.
Key Responsibilities:
- Train and fine-tune LLMs for task performance and alignment using SFT and alignment methods (RLHF, PPO, GRPO), and validate with robust evaluation practices.
- Build production-grade RAG systems including ingestion, chunking, embedding, vector search, reranking, grounding, and evaluation to reduce hallucinations.
- Develop agentic AI workflows using LangChain and LangGraph, including tool calling, multi-step planning, memory, guardrails, and observability.
- Deploy and serve models locally on GPUs using vLLM and Ollama, optimizing throughput, latency, batching, and KV cache behavior.
- Productionize distributed inference and services using Ray and Kubernetes (deployments, autoscaling, rolling updates, reliability).
- Build end-to-end voice agents using LiveKit (streaming STT, LLM orchestration, TTS, turn-taking, and real-time session handling).
- Collaborate with product and engineering teams to define requirements, success metrics, and deliver production-ready features with documentation and tests.
Knowledge/Skills/Abilities:
- Strong Python skills with practical experience in PyTorch and modern LLM tooling.
- Familiarity with distributed GPU inference concepts (tensor parallelism, caching, throughput tuning).
- Working knowledge of CI/CD, monitoring, logging, and tracing for AI services.
- Working knowledge of alignment techniques and workflows such as RLHF, PPO, and GRPO.
Terms & Conditions:
- Candidates are required to attach scanned copies of their documents (Academics /Professional).
- Last education certificate/degree must be attested/verified by HEC.
Candidates may be considered ineligible for the post due to any of the following reasons:
- 3rd Div in academic career / weak academic profile.
- NUST employees with less than one year of service with NUST and / or absence of NOC from Head of Institution.
- In process of pursuing a required degree.
- Medically unfit.
- Only selected candidates will be contacted and issued offer letter.
- Candidates serving in Govt departments, Armed forces may apply through their respective parent department / organizations.
- Late / incomplete applications will be ignored.
- Only short-listed candidates will be considered / called for test / interview and no TA / DA will be admissible.
- NUST reserves the right to cancel, modify / terminate the recruitment programme due to any reason, without notice, at any time.
© 2025 Qureos. All rights reserved.