About the role
Shodh AI is hiring an AI/ML Engineer to help build enterprise AI applications using open-weight language models. You will work on document understanding, structured information extraction, model fine-tuning and evaluation. This is a hands-on engineering role. You will own clearly scoped tasks, supported by our senior technical team for architecture, complex troubleshooting and deployment decisions. What you’ll do
• Prepare, clean and validate datasets, including synthetic training examples.
• Adapt language models using supervised fine-tuning and techniques such as LoRA/QLoRA.
• Develop prompts and pipelines for extracting business rules and producing structured outputs.
• Build tests to evaluate correctness, consistency and handling of ambiguous or missing information.
• Integrate models with application APIs alongside backend engineers.
• Help package and deploy models in GPU-backed, on-premises and offline environments.
• Document experiments, results and known limitations. What we’re looking for




• 1–2 years of relevant experience in AI/ML engineering.
• Strong Python programming and debugging skills.
• Practical experience with PyTorch and Hugging Face Transformers.
• Hands-on experience building an LLM application, fine-tuning a model or deploying an inference service.
• Understanding of training, validation and test datasets.
• Familiarity with Git and Linux.
• Transparent communication, curiosity and a willingness to learn. Good to have
• Exposure to PEFT, LoRA/QLoRA, TRL or vLLM.
• Experience with document processing, JSON outputs and schema validation.
• Familiarity with Docker, quantization or GPU inference.

How to apply
• Apply through LinkedIn with your CV and a relevant project link or brief project description. Please explain what you personally built. You do not need to share confidential code or data.

📌 AI/ML Engineer — Generative AI (Jaipur)
🏢 Shodh AI
📍 Jaipur