About CIBC India:
CIBC India is a technology and operations hub in Hyderabad, where you'll be part of our highly engaged and global team, and play a central role in supporting our growth. Whether you're driving innovation in digital banking or streamlining client onboarding, you'll be part of a culture recognized for excellence and investment in people. At CIBC India, people and progress are at the centre of what we do: you'll develop skills, collaborate with industry leaders, and see your ideas realized in a culture that values progress and belonging. We provide cutting-edge resources, ongoing learning, and a supportive environment where people come first. If you're ready to create meaningful change and build your future, CIBC India is where your ambition meets opportunity.
What You'll Be Doing (Position Summary):
The Director Agentic Engineer is responsible for designing, building, and optimizing multi-agent AI systems to support front-office, risk, and operations teams across Capital markets. This role leverages modern agent orchestration frameworks, such as LangGraph, Pydantic.AI, and LangChain, to develop core planning and orchestration logic, tool integrations, retrieval pipelines, and agent lifecycle management. The Agentic Engineer plays a key role in advancing the next-generation Agentic AI platform, ensuring robust, scalable, and secure autonomous workflows for business-critical functions.
At CIBC India we enable the work environment most optimal for you to thrive in your role. Details on your work arrangement (including on-site and remote work) will be discussed at the time of your interview.
How You'll Succeed (Responsibilities):
Multi-Agent System Design & Implementation:
Design and implement multi-agent workflows, planner agents, tool-calling logic, and orchestration graphs using agentic frameworks (LangGraph preferred).Build production-ready agents that interact with risk, PnL, RFQ, pricing, market-data, trade-capture, and reference-data systems.
Integration & Tooling:
Develop robust tool wrappers, MCP-based connectors, and internal service APIs for agent access.Implement retrieval strategies, including RAG, hybrid search, semantic retrieval, tabular retrieval, and agent memory patterns.
Performance Optimization & Monitoring:
Optimize cost, latency, and reliability of agent executions across distributed compute environments.Implement Open Telemetry-based agent logs, traces, and evaluation pipelines.
Security & Governance:
Collaborate with platform engineering to define and implement security, entitlements, ABAC/RBAC, and sandboxing for agents.Contribute to agent governance processes, including versioning, approvals, evaluation, and testing.
Business Collaboration & Prototyping:
Work with business users to prototype and deliver new autonomous workflows (e.g., morning briefings, monitoring agents, reconciliation agents).
Software Development:
Write production-quality Python code with strong type integrity, utilizing Pydantic Models.
Experience: 12+ years of experience
Who You Are (skills/qualifications):
Must Have Skills:
Strong experience with agentic frameworks such as LangGraph, Pydantic.AI, LangChain, or equivalents (AutoGen, Semantic Kernel, DSPy, CrewAI, etc.).Advanced proficiency in Python, including async programming, concurrency, and distributed execution.Experience integrating LLMs (OpenAI, Azure OpenAI, Anthropic, local models, vLLM).Proven ability to build tool functions, structured prompts, agent state machines, and execution graphs.Practical expertise with RAG/retrieval architecture, vector stores, embeddings, and text ranking.Understanding of MCP(Model Context Protocol or event-driven agent meshes.Familiarity with tabular and unstructured retrieval, hybrid pipelines, SQL connectors, semantic layers, or metadata systems.Experience working with structured financial data (e.g., derivatives, risk, limits, onboarding, trade flows) is an asset.Implement Open Telemetry-based agent logs, traces, and evaluation pipelines.
Good to Have:
Experience with Rust, TypeScript, or .NET for high-performance agent tooling.Experience with Databricks, Delta Lake, Spark, or serverless compute platforms.Experience in capital markets, particularly derivatives.Experience deploying agent workloads on Kubernetes or serverless execution engines.
What CIBC India Offers:
At CIBC India, your goals are a priority. We start with your strengths and ambitions and strive to create opportunities to tap into your potential. We aspire to give you a career that goes well beyond your compensation.
We work to recognize you in meaningful, personalized ways including competitive fixed pay, performance-based incentive programs, a family-first benefits program, employer contributions for your retirement, attractive time off and leave offerings, wellbeing support, and MomentMakers, our social, points-based recognition program.Our spaces and technological toolkit will make it simple to bring together great minds to create innovative solutions that make a difference for our clients worldwide.We cultivate a culture where you can express your ambition through initiatives like Purpose Day, a paid day off dedicated for you to use to invest in your growth and development.
Privacy:
By proceeding with the creation of a profile and submitting an application in response to this job posting, you acknowledge and agree that your personal data will be collected, used, retained and disclosed by CIBC India and its affiliates, as described in the Candidate Privacy Statement.
Read Less
