At CommBank, we never lose sight of the role we play in other people’s financial wellbeing. Our focus is to help people and businesses move forward to progress. To make the right financial decisions and achieve their dreams, targets, and aspirations. Regardless of where you work within our organisation, your initiative, talent, ideas, and energy all contribute to the impact that we can make with our work. Together we can achieve great things.

Job Title:  Data Scientist

Location:  Bangalore

Business & Team: BB Advanced Analytics and Artificial Intelligence COE

Impact & contribution:

As a Data Scientist, you will be instrumental in pioneering Gen AI and multi-agentic systems at scale within CommBank. You will architect, build, and operationalize advanced generative AI solutions—leveraging large language models (LLMs), collaborative agentic frameworks, and state-of-the-art toolchains. You will drive innovation, helping set the organizational strategy for advanced AI, multi-agent collaboration, and responsible next-gen model deployment.

Roles & Responsibilities:

• Gen AI Solution Development:
  Lead end-to-end development, fine-tuning, and evaluation of state-of-the-art LLMs and multi-modal generative models (e.g., transformers, GANs, VAEs, Diffusion Models) tailored for financial domains.

• Multi-Agentic System Engineering:
  Architect, implement, and optimize multi-agent systems, enabling swarms of AI agents (utilizing frameworks like Lang chain, Lang graph, and MCP) to dynamically collaborate, chain, reason, critique, and autonomously execute tasks.

• LLM-Backed Application Design:
  Develop robust, scalable GenAI-powered APIs and agent workflows using Fast API, Semantic Kernel, and orchestration tools. Integrate observability and evaluation using Lang fuse for tracing, analytics, and prompt/response feedback loops.

• Guardrails & Responsible AI:
  Employ frameworks like Guardrails AI to enforce robust safety, compliance, and reliability in LLM deployments. Establish programmatic checks for prompt injections, hallucinations, and output boundaries.

• Enterprise-Grade Deployment:
  Productionize and manage at-scale Gen AI and agent systems with cloud infrastructure (GCP/AWS/Azure), utilizing model optimization (quantization, pruning, knowledge distillation) for latency/throughput trade-offs.

• Toolchain Innovation:
  Leverage and contribute to open-source projects in the Gen AI ecosystem (e.g., Lang Chain, Lang Graph, Semantic Kernel, Lang fuse, Hugging face, Fast API). Continuously experiment with emerging frameworks and research.

• Stakeholder Collaboration:
  Partner with product, engineering, and business teams to define high-impact use cases for Gen AI and agentic automation; communicate actionable technical strategies and drive proof-of-value experiments into production.

• Mentorship & Thought Leadership:
  Guide junior team members in best practices for Gen AI, prompt engineering, agentic orchestration, responsible deployment, and continuous learning. Represent CommBank in the broader AI community through papers, patents, talks, and open-source.

Essential Skills:

• 6+ years of hands-on experience in Machine Learning, Deep Learning, or Generative AI domains, including practical expertise with LLMs, multi-agent frameworks, and prompt engineering.

• Proficient in building and scaling multi-agent AI systems using Lang Chain, Lang Graph, Semantic Kernel, MCP, or similar agentic orchestration tools.

• Advanced experience developing and deploying Gen AI APIs using Fast API; operational familiarity with Lang fuse for LLM evaluation, tracing, and error analytics.

• Demonstrated ability to apply Guardrails to enforce model safety, explainability, and compliance in production environments.

• Experience with transformer architectures (BERT/GPT, etc.), fine-tuning LLMs, and model optimization (distillation/quantization/pruning).

• Strong software engineering background (Python), with experience in enterprise-grade codebases and cloud-native AI deployments.

• Experience integrating open and commercial LLM APIs and building retrieval-augmented generation (RAG) pipelines.

• Exposure to agent-based reinforcement learning, agent simulation, and swarm-based collaborative AI.

• Familiarity with robust experimentation using tools like Lang Smith, GitHub Copilot, and experiment tracking systems.

• Proven track record of driving Gen AI innovation and adoption in cross-functional teams.

• Papers, patents, or open-source contributions to the Gen AI/LLM/Agentic AI ecosystem.

• Experience with financial services or regulated industries for secure and responsible deployment of AI.

Education Qualifications: Bachelor’s or master’s degree in computer science, Engineering, Information Technology.