Senior ML Engineer

About Shodh AI

Shodh AI is an AI technology company that enables businesses to build custom AI models tailored to their unique needs. We are building platforms that simplify the process of creating Enterprise Language Models (ELMs) that comprehend and utilize your company's language. Our models are versatile, deployable on any cloud service or directly on your servers, and accessible via an API. We have our own H100 clusters and are currently building tools for various unicorns, companies, as well as DRDO. We were co-founded by a former Microsoft Cambridge engineer and a leading AI scientist from India, and our mission is to revolutionize businesses with AI that truly understands their operations.

Job Description

We are urgently seeking an experienced and highly motivated Senior Machine Learning Engineer with a specialization in Reinforcement Learning or Natural Language Processing. This role is in collaboration with the Defence Research and Development Organisation (DRDO) and will be based in their Bangalore office. The ideal candidate will have at least two years of hands-on experience with deep learning frameworks such as PyTorch and Lightning. Proficiency in distributed training (DDP), CUDA, NCCL, and high-performance computing (HPC) is essential. Additionally, the candidate should possess extensive knowledge of RL or NLP with practical experience in frameworks like Hugging Face, torch, stable baselines, or nltk. Knowledge of tools such as git, ml-ops, Kubernetes, Docker, and enroot is preferred.

Responsibilities

• Develop and deploy cutting-edge machine learning solutions utilizing LLMs and deep learning networks to address client needs.
• Implement and test different model architectures, leveraging deep learning frameworks such as PyTorch and Lightning.
• Utilize distributed training techniques (DDP) and tools (CUDA, NCCL) for efficient model training on HPC clusters.
• Drive innovation in the ML market by creating unique and alternative solutions.
• Collaborate with teams to develop and implement efficient and innovative machine learning solutions tailored to client needs.
• Employ Kubernetes and Docker for model deployment and management.
• Use evaluation strategies to assess and improve performance.

Requirements

• Minimum of 2 years of hands-on experience with RL/NLP.
• Proficiency in deep learning frameworks: PyTorch, Lightning.
• Strong understanding of distributed processing systems and tools: DDP, CUDA, NCCL, HPC clusters.
• Extensive knowledge of NLP literature and frameworks like Hugging Face, llama.cpp, and GPT.
• Extensive knowledge of RL literature and frameworks like Hugging Face, stable baselines, and torchrl.
• Experience with Kubernetes and Docker for model deployment.
• Familiarity with evaluation strategies for learning models.
• Basic Linux and Python skills to efficiently implement standard practices in project collaboration.

Qualifications

• Demonstrated ability to work independently and in a team environment.
• Excellent problem-solving and communication skills.
• Passion for staying at the forefront of technology and exploring new advancements in the field.

Location: DRDO, Bangalore

Date of Joining: Immediate — We are looking to fill this position as soon as possible.
Candidates are expected to join within 2 weeks of receiving the offer letter.

Compensation: Competitive salary (~8-16 LPA, based in India) and long-term ESOP options.

Term of Employment: We aim to have you as a long-term member of our growth journey.