--

We are looking for a passionate AI Researcher to join our team. You will work at the intersection of cutting-edge AI research and product engineering—designing, evaluating, and deploying generative AI (GenAI) systems that are both reliable and impactful. This role blends fundamental research, model evaluation, and practical software engineering to push forward the next generation of intelligent applications.

What will your typical day look like?

Responsibilities
•    Collaborate with product managers, engineers, and stakeholders to design AI-driven solutions that meet technical and business requirements. 
•    Research, prototype, and develop generative AI applications by combining non-deterministic LLMs with deterministic software engineering techniques.  
•    Build evaluation frameworks and benchmarks to measure model quality, reliability, and business impact.  
•    Generate regular reports on model accuracy, drift, and performance.  
•    Debug, optimize, and enhance GenAI applications using prompt engineering, reinforcement learning, fine-tuning, and software engineering best practices.  
•    Train and fine-tune large language models using Hugging Face Transformers.  
•    Apply reinforcement learning fine-tuning techniques using Hugging Face TRL (Transformers Reinforcement Learning).  
•    Manage training workflows with experiment tracking tools and distributed training accelerators (DeepSpeed, Accelerate, FSDP).  
•    Run and optimize multi-GPU training and inference, leveraging vLLM for high-throughput, low-latency serving.  
•    Contribute to the design of scalable MLOps/DevOps pipelines for model deployment, monitoring, and continuous training.  
•    Ensure compliance with data privacy, security, and responsible AI guidelines when handling training or test datasets. 
•    Stay current with emerging research in LLMs, RLHF/RLAIF, multimodal AI, and generative models; apply findings to improve our systems.  
•    Author technical documentation and contribute to publications, patents, or open-source projects where applicable. 

About the team

Preferred Tech Stack
Candidates with experience in the following tools and frameworks will be strongly preferred:
•    Transformers (Hugging Face) for model training, fine-tuning, and inference  
•    TRL (Transformers Reinforcement Learning) for RL-based fine-tuning (PPO, DPO, GRPO, RLAIF)  
•    DeepSpeed, Accelerate, or FSDP for multi-GPU and distributed training  
•    vLLM for optimized inference and serving of large models  
•    Weights & Biases (W&B) or MLflow for experiment tracking and reproducibility  
•    LangChain, AutoGen (A2A), or MCP for GenAI application development  
•    PyTorch as the primary deep learning framework  

Enough about us, let’s talk about you

Qualifications
•    3+ years experience in machine learning engineering, data engineering, or applied research (industry or academic).  
•    Strong programming skills in Python and experience with frameworks such as PyTorch, TensorFlow, JAX.  
•    Hands-on experience with Hugging Face Transformers for pretraining, fine-tuning, or inference.  
•    Experience with Hugging Face TRL for reinforcement learning fine-tuning (e.g., PPO, DPO, GRPO, RLAIF).  
•    Practical experience managing multi-GPU training and distributed training at scale using DeepSpeed, Accelerate, or FSDP.  
•    Experience running inference on large models using vLLM or similar optimized serving frameworks.  
•    Familiarity with experiment tracking and reproducibility tools (e.g., W&B, MLflow).  
•    Knowledge of MLOps practices including continuous training, continuous monitoring, and model lifecycle management.  
•    Experience with GenAI frameworks such as LangChain, AutoGen (A2A), or MCP.  
•    Demonstrated ability to write clean, maintainable, production-ready code.  
•    Experience building or supporting cloud-based AI systems (GCP, AWS, or Azure; certifications preferred).  
•    Strong grasp of reinforcement learning, NLP, and/or generative modeling (transformers, diffusion, RAG, etc.).  
•    Track record of research contributions (papers, patents, open-source projects) is a plus.  
Nice-to-Have
•    Experience with reinforcement learning from human/AI feedback (RLHF/RLAIF).  
•    Contributions to open-source AI frameworks.  
•    Familiarity with scaling laws, evaluation metrics, and benchmarking large models.  
•    Interest in pushing the boundaries of trustworthy, explainable, and safe AI.

Total Rewards

The salary range for this position is $72,000 - $138,000, and individuals may be eligible to participate in our bonus program. Deloitte is fair and competitive when it comes to the salaries of our people. We regularly benchmark across a variety of positions, industries, sectors, targets, and levels. Our approach is grounded on recognizing people's unique strengths and contributions and rewarding the value that they deliver.

Our Total Rewards Package extends well beyond traditional compensation and benefit programs and is designed to recognize employee contributions, encourage personal wellness, and support firm growth.  Along with a competitive base salary and variable pay opportunities, we offer a wide array of initiatives that differentiate us as a people-first organization. On top of our regular paid vacation days, some examples include: $4,000 per year for mental health support benefits, a $1,300 flexible benefit spending account, firm-wide closures known as "Deloitte Days", dedicated days of for learning (known as Development and Innovation Days), flexible work arrangements and a hybrid work structure.