Software Engineer AI Platform

BUSINESS UNIT DESCRIPTION:

The Capital Markets Technology Team at CIBC is responsible for delivering advanced technology solutions that drive the success of the bank's Capital Markets business. This team encompasses specialized groups, including Derivatives Technology, Capital Markets Data Analytics, Electronic Trading, Fixed Income, FX, and Risk Management Software, Equities Technology etc each playing a vital role in supporting efficient, secure, and scalable trading and analytics platforms. By developing and maintaining robust systems for derivativesincluding sophisticated risk management software essential for monitoring, analyzing, and mitigating derivatives riskenabling high-performance electronic trading, providing comprehensive technology for fixed income and foreign exchange trading, and delivering actionable insights through advanced data analytics, the team ensures operational excellence, regulatory compliance, and superior client service. Through continuous innovation and alignment with strategic business objectives, the Capital Markets Technology Team empowers CIBC to manage risk effectively, respond to market opportunities, and deliver exceptional value to clients.

JOB PURPOSE:

Position summary:

- The team: You will join a senior team building CIBC's agentic AI platform for Capital Markets. As a Software Engineer on the AI Platform, you will build, test, and operate the reusable services that give AI agents and applications reliable, secure access to structured and unstructured data.

- The role: A hands-on individual contributor role, working alongside senior engineers, with a clear path to a senior engineering role as you grow. You will work across ingestion, search, retrieval APIs, metadata, caching, and evaluation, using Python, FastAPI, Qdrant, Azure AI Search, Databricks, LangGraph, Pydantic, and MLflow.

- Who we are looking for: Engineers with a solid track record of shipping production software and practical experience with search and retrieval systems.

- Growth: As the platform grows, the role offers scope to expand into semantic layers over risk and PnL data, knowledge graphs, real-time data flows, and agentic workflows.

Responsibilities:

- Delivery: Write, review, test, and release production code every day at a high pace, without compromising quality.

- Ownership: Own features and components end to end, from design to production, with guidance from senior engineers on larger design decisions.

- AI-assisted development: Use AI coding assistants to accelerate delivery, while critically reviewing generated code for correctness, security, and maintainability.

- Ingestion and retrieval services: Build reusable ingestion and retrieval services for unstructured documents and structured data, from parsing and indexing through to context assembly.

- APIs and agent tools: Define clear APIs, metadata schemas, and structured outputs, and expose retrieval as tools that AI agents can call.

- Retrieval quality: Help measure and improve retrieval quality (including precision, recall, and hallucination reduction), latency, and cost against clear targets, using techniques such as hybrid search, reranking, and caching.

- Traceability: Ensure retrieval is accurate, context-aware, and auditable, so answers can be traced back to their sources.

- Reliability: Help run services reliably in production with automated testing, CI/CD, containerized deployment, observability, and failure recovery.

- Security and governance: Build access controls, data lineage, and safe handling of sensitive data into every retrieval path.

- Collaboration: Work closely with platform, data, security, and front-end teams across India and the UK, including some overlap with UK working hours.

- Engineering standards: Document your designs and decisions clearly, and follow and contribute to the team's engineering standards.

KNOWLEDGE AND SKILLS:

You bring 7+ years of software engineering experience. You have personally built and shipped production software, with well-tested code and hands-on experience supporting it in production. Experience with a high-volume or business-critical system is a plus. Be ready to explain your own technical contribution, the trade-offs you made, and the outcomes.

Must-have skills:

- Python software engineering: Building maintainable Python services and APIs (e.g., FastAPI, Pydantic); typing and error handling; writing effective tests with pytest; explaining design choices clearly.

- Search, retrieval, and RAG: Fundamentals of indexing, chunking, embeddings, vector and keyword search, metadata filtering, and reranking; how to check whether retrieval results are good and why they might be wrong.

- Production engineering: Linux and shell, Docker, CI/CD; reading logs and metrics; debugging and fixing production issues; handling timeouts and retries.

- AI-assisted software development: Daily use of AI coding agents (e.g., GitHub Copilot, Claude Code) to work faster, with the judgement to review and test generated code before it ships.

- LLM application engineering: Calling LLM and embedding APIs; prompt design, structured outputs, and tool/function calling; awareness of cost, latency, and hallucination risks.

- Data handling: Working with data from multiple sources; schemas, metadata, and access controls.

- Learning agility: The ability to learn unfamiliar technologies and domains quickly.

Good-to-have skills:

- RAG in production: Hands-on experience running RAG in production, including retrieval evaluation.

- Our stack: Experience with FastAPI, Qdrant, Azure AI Search, Databricks, LangGraph, Pydantic, or MLflow.

- Azure: Building on Azure, including containers, Kubernetes, and identity-based access.

- Document parsing: Extracting content from PDFs, tables, and scanned documents (OCR).

- Semantic models and knowledge graphs: Exposure to semantic models, metadata systems, or knowledge graphs.

- Agent platforms: Agent frameworks and multi-agent workflows, or exposing platform services to AI agents as reusable tools.

- Mentoring: Supporting or mentoring other engineers.

- Capital markets: Familiarity with capital-markets data (such as trades, curves, surfaces, risk, and PnL), or experience delivering software in another complex, regulated domain.

Preferred: Bachelor's degree or higher in Computer Science or a related field, or equivalent practical experience.

COMPLIANCE REQUIREMENTS/RESPONSIBILITIES:

As an employee of CIBC, the incumbent must comply with all applicable CIBC and Line of Business policies, standards, guidelines and controls.

WORKING CONDITIONS:

This role operates within a normal office environment with little exposure to adverse working conditions. 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.

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 theCandidate Privacy Statement.