Senior Staff AI Engineer
JazzX AIBangaloreOn-Site
Needs a free Superjobs account. You'll finish on JazzX AI's Greenhouse page.
Skill score
Summary
The Senior Staff AI Engineer leads the architecture and production deployment of reinforcement learning systems for JazzX AI's enterprise AI platform, including training pipelines, simulation, multi-agent RL and model serving. The role suits a senior engineer with deep RL and LLM experience who can also mentor engineers and bridge research with production.
- Function
- Data Science
- Estimated salary
- Est. ₹25–38 LPA
- Estimated for a staff-level role, ~10 yrs experience, India. Not published by the employer.
- Work mode and location
- On-Site, Bangalore
- Experience
- 10+ yrs
Job description
Overview
We are seeking an experienced AI Engineer with deep expertise in Reinforcement Learning (RL) to join our team as a Senior Staff Architect. In this role, you will be responsible for shaping the vision, architecture, and technical execution of RL-driven AI reasoning models and systems that power next-generation enterprise AGI platform.
You will lead the design, development, and optimization of cutting-edge RL solutions, from experimentation and simulation through production deployment. This includes building scalable training architectures, architecting multi-agent and hierarchical RL frameworks, and ensuring that the RL systems are resilient, efficient, explainable and safe.
As a senior technical leader, you will partner with cross-functional teams—including product, core platform engineering, and research—to define architectural best practices, establish governance standards, and enable seamless integration of RL into our broader AGI platform. You will also drive innovation by exploring novel RL techniques, mentoring engineers and researchers, and ensuring the RL infrastructure can scale to support high-throughput training and real-world scenarios and enterprise use cases end to end.
Ultimately, your work will be critical in bridging research and production, ensuring that the latest RL advancements translate into reliable, impactful, and enterprise-ready AI solutions.
Roles and responsibilities
- Architecture & Design: Define and drive the end-to-end architecture for reinforcement learning–based systems, including training pipelines, simulation environments, reward shaping, and model serving.
- Research & Development: Apply cutting-edge RL techniques (policy optimization, model-based RL, hierarchical RL, multi-agent RL, etc) to solve complex enterprise problems.
- Scalability & Infrastructure: Design distributed training systems, leverage cloud-native infrastructure, and optimize for performance, reproducibility, and cost-efficiency.
- Leadership & Mentorship: Provide technical leadership to AI engineers and researchers; mentor junior team members; review designs and code with a focus on scalability, robustness, and clarity.
- Collaboration: Partner with product, data, and platform teams to align RL solutions with strategic business goals and integrate them into production systems.
- Evaluation & Monitoring: Define frameworks for benchmarking, continuous evaluation, and feedback-driven improvements in deployed RL models.
- Compliance & Safety: Ensure RL systems align with ethical AI practices, safety constraints, and regulatory standards for enterprises.
Skills required
Top skills
- Test-time compute optimization
- Policy optimization
- Multi-agent RL
- PPO
All skills (10)
- SAC
- RLVR
- GRPO
- LLM fine-tuning
- Distributed training
- PyTorch
- Kubernetes
- Python
- Reinforcement learning
- Simulation frameworks
Qualifications
- 10+ years of experience in AI/ML engineering, including at least 5 years specializing in reinforcement learning research and production systems.
- Demonstrated success in designing and deploying large-scale RL architectures in enterprise environments.
- Deep expertise in reinforcement learning algorithms, including on-policy (PPO, A3C) and off-policy (SAC, DDPG) methods, along with hands-on work in simulation frameworks (e.g., OpenAI Gym, Isaac Gym, PettingZoo, MuJoCo).
- Practical experience with multi-agent reinforcement learning (MARL), including coordination strategies for complex environments.
- Strong proficiency in Reinforcement Learning with Verifiable Rewards (RLVR) and GRPO-like policy optimization approaches, applying reinforcement learning principles both rigorously and pragmatically.
- Experience with test-time compute optimization techniques, including inference-time search, chain-of-thought reasoning, and adaptive computation strategies for improving model performance during deployment.
- Proven ability in large language model (LLM) training and fine-tuning, across both supervised and reinforcement learning–driven techniques.
- Advanced software engineering skills in Python, C++, or Java, with deep expertise in ML frameworks such as TensorFlow, PyTorch, JAX, or Ray RLlib.
- Hands-on experience with distributed training infrastructure (Kubernetes, GPU/TPU clusters, and cloud ML platforms).
- Excellent communication, collaboration, and leadership skills, with experience working across multidisciplinary teams.
Nice to have
- PhD in Computer Science, Machine Learning, Robotics, or related field.
- Experience leading enterprise AI adoption and guiding organizational strategy for RL powered systems.
- Contributions to open-source RL frameworks or publications in top-tier conferences (NeurIPS, AISTATS, ICML, ICLR, AAAI).
- Background in safety, alignment, or explainability of RL agents.
Other details
We offer a competitive salary ranging from USD 207,000 to USD 290,000, along with equity options and an attractive benefits package, including health, dental, and vision insurance, flexible working arrangements, and more.
Learn more about JazzX AI:Website: https://jazzx.ai/?utm_source=chatgpt.comLinkedIn: https://www.linkedin.com/company/jazzx-ai?utm_source=chatgpt.com
Learn more about SAIGroup:Website: https://saigroup.ai/
Company info
About JazzX AI
JazzX AI is defining the future of enterprise work—by building AI-native digital workers that actually get the job done. JazzX AI transforms messy enterprise reality into institutional intelligence: governed digital workers that capture expert judgment, make every decision explainable, and continuously improve through real-world execution. The result is faster decisions,