Senior ML/AI Engineer

Syren Cloud Inc.; Hyderabad; On-site; INR 1,800,000 - 3,000,000; Full time

Job summary

Syren Cloud Inc. is seeking an experienced ML/AI engineer to build, train, and deploy cutting-edge models with a focus on LLM fine-tuning and scalable inference pipelines.

You will design data pipelines, implement thorough evaluation, and collaborate with platform engineers to ensure robust production serving. The role requires strong Python and PyTorch skills, hands-on LLM fine-tuning experience, and a solid grounding in MLOps and distributed multi-GPU training.

Qualifications

• 3+ years of ML/AI engineering experience.
• Strong Python and PyTorch background.
• Hands-on LLM fine-tuning (LoRA/PEFT) experience.
• Experience building inference/serving pipelines.
• Familiar with experiment tracking, versioning, and MLOps.
• Distributed training with multi-GPU setups.
• Publications or notable open-source ML work.
• Experience with embeddings, retrieval, and evaluation frameworks.

Responsibilities

• Build, train, and deploy ML models — with focus on LLM fine-tuning and high-performance inference pipelines.
• Train, fine-tune, and evaluate models — including LLMs via full fine-tuning, LoRA/PEFT, and instruction tuning.
• Build and optimize inference pipelines for latency, throughput, and cost (batching, quantization, caching).
• Design data pipelines and datasets for training and evaluation; ensure data quality and reproducibility.
• Implement rigorous offline and online evaluation, benchmarking, and regression testing for models.
• Partner with platform engineers to deploy models to production with monitoring and rollback.
• Stay current with model architectures and translate research into practical improvements.

Skills

Python

PyTorch

Transformer

LLM fine-tuning

Inference pipelines

MLOps basics

Distributed training

Open-source ML

Embeddings retrieval

Experiment tracking

Job description

Build, train, and deploy machine learning models — with a focus on LLM fine-tuning and high-performance inference pipelines. You own models from experimentation through to reliable production serving.

• Train, fine-tune, and evaluate models — including LLMs via full fine-tuning, LoRA/PEFT, and instruction tuning.
• Build and optimize inference pipelines for latency, throughput, and cost (batching, quantization, caching).
• Design data pipelines and datasets for training and evaluation; ensure data quality and reproducibility.
• Implement rigorous offline and online evaluation, benchmarking, and regression testing for models.
• Partner with platform engineers to deploy models to production with monitoring and rollback.
• Stay current with model architectures and translate research into practical improvements.

Required Skills

• 3+ years in ML/AI engineering; proven track record shipping models to production.
• Strong Python and deep learning with PyTorch; solid understanding of the Transformer architecture.
• Hands‑on LLM fine‑tuning experience (LoRA/PEFT, quantization, or full fine‑tuning).
• Experience building inference/serving pipelines and measuring model performance.
• Comfortable with experiment tracking, versioning, and MLOps fundamentals.
• Distributed training experience (multi‑GPU, DeepSpeed, FSDP).
• Publications, competition results, or notable open-source ML work.
• Experience with retrieval, embeddings, and evaluation frameworks.

Core Stack: