AI Engineer LLM Fine-Tuning & Reasoning Systems

ZySec AI; Hyderabad; On-site; INR 1,200,000 - 2,100,000; Full time

Job summary

ZySec AI in Hyderabad is seeking an experienced AI Engineer to push LLM fine-tuning and reasoning systems. You will optimize LLMs, experiment with quantization, and build efficient reasoning architectures for production deployment.

Collaborate with data engineering, infra, and applied research teams to bring ideas from paper to production. 3+ years in deep learning and LLMs required; strong PyTorch internals and open-source contributions preferred.

Qualifications

• 3+ years of hands-on DL experience with LLMs and deep learning.
• Strong understanding of PyTorch internals and memory profiling.
• Proven track record in fine-tuning LLMs on real-world tasks.
• Ability to design and run standardized evals (MMLU, GSM8K, HELM).
• Deep familiarity with quantization techniques and low-bit inference.
• Working knowledge of Hugging Face ecosystem and public repos.
• Experience in training multi-modal models is a plus.

Responsibilities

• Fine-tune Large Language Models on custom datasets for specialized tasks.
• Design benchmarking pipelines across accuracy, speed, and efficiency.
• Implement quantization, pruning, and distillation for deployment-readiness.
• Evaluate retrieval-augmented generation pipelines and reasoning agents.
• Contribute to SOTA architectures for multi-hop and multimodal reasoning.
• Collaborate with data engineering, infra, and research teams end-to-end.
• Own experiments, ablations, and performance dashboards.

Skills

LLMs

PyTorch internals

LLM fine-tuning

Benchmarking

Quantization

Hugging Face

Multimodal training

Open-source contributions

Tools

Transformers

Accelerate

Datasets

Evaluate

Job description

AI Engineer LLM Fine-Tuning & Reasoning Systems

About the job AI Engineer LLM Fine-Tuning & Reasoning Systems

We're building the future of Autonomous Data Intelligence at CyberPod AI and were looking for a deeply technical, hands-on AI Engineer to push the boundaries of whats possible with Large Language Models (LLMs).

This role is for someone whos already been in the trenches : fine-tuned foundation models, experimented with quantization and performance tuning, and knows PyTorch inside out . If youre passionate about optimizing LLMs, crafting efficient reasoning architectures, and contributing to open-source communities like Hugging Face , this is your playground .

What You'll Do

• Fine-tune Large Language Models (LLMs) on custom datasets for specialized reasoning tasks.
• Design and run benchmarking pipelines across accuracy, speed, token throughput, and energy efficiency.
• Implement quantization, pruning, and distillation techniques for model compression and deployment readiness.
• Evaluate and extend agentic RAG (Retrieval-Augmented Generation) pipelines and reasoning agents.
• Contribute to SOTA model architectures for multi-hop, temporal, and multimodal reasoning.
• Collaborate closely with the data engineering, infra, and applied research teams to bring ideas from paper to production.
• Own and drive experiments, ablations, and performance dashboards end-to-end.

Requirements

• 3+ years of hands-on experience working with deep learning and large models, particularly LLMs.
• Strong understanding of PyTorch internals: autograd, memory profiling, efficient dataloaders, mixed precision.
• Proven track record in fine-tuning LLMs (e.g., LLaMA, Falcon, Mistral, Open LLaMA, T5, etc.) on real-world use cases.
• Benchmarking skills: can run standardized evals (e.g., MMLU, GSM8K, HELM, TruthfulQA) and interpret metrics.
• Deep familiarity with quantization techniques: GPTQ, AWQ, QLoRA, bitsandbytes, and low-bit inference.
• Working knowledge of Hugging Face ecosystem (Transformers, Accelerate, Datasets, Evaluate).
• Active Hugging Face profile with at least one public model/repo published.
• Experience in training and optimizing multi-modal models (vision-language/audio) is a big plus.
• Published work (arXiv, GitHub, blogs) or open-source contributions preferred.

If you are passionate about AI and want to be a part of a dynamic and innovative team, then ZySec AI is the perfect place for you.