AI ML Gen AI Engineer (Banking Compliance)

Key Responsibilities

AI Solution Design & Engineering

● Partner with product and business teams to translate banking problems (fraud, credit risk,

customer operations, compliance) into practical AI solutions.

● Determine when to apply traditional ML versus LLM/GenAI approaches, evaluating trade-offs

across accuracy, latency, cost, and regulatory constraints.

● Design and implement end-to-end AI systems including data pipelines, feature engineering,

model integration, and API-based services.

● Build and evolve agentic and RAG (Retrieval-Augmented Generation) architectures using

frameworks such as LangChain or LangGraph.

Production Engineering & Delivery

● Build production-ready AI services with robust error handling, fallback mechanisms, guardrails,

and observability (logging, metrics, tracing).

● Implement AI safety controls including input validation, prompt injection mitigation,

configurable policies, and kill-switch mechanisms.

● Optimise AI systems for performance, latency, and cost — particularly important for high

volume banking workloads.

● Transition PoCs and prototypes into hardened production systems through refactoring, testing,

and rigorous deployment practices.

● Work with SQL, NoSQL, and vector databases (e.g., PostgreSQL, MongoDB, ChromaDB) to

support data-intensive AI applications.

ML & Generative AI

● Apply supervised and unsupervised ML techniques to banking use cases such as classification,

anomaly detection, and recommendation.

● Build and integrate LLM-based solutions using models such as OpenAI, Claude, Gemini, Llama, or

equivalent.

● Apply prompt engineering, evaluation techniques, and iterative optimisation to improve GenAI

output quality.

● Develop tool-based and agentic workflows, including multi-agent systems for complex, multi

step banking processes.

Collaboration & Communication

● Collaborate with platform, cloud, and infrastructure teams to ensure reliable deployment and

operations.

● Clearly articulate trade-offs (ML vs. LLM, build vs. buy, speed vs. robustness) to both technical

and non-technical stakeholders.

● Uphold strong software engineering practices: code quality, documentation, version control, and

CI/CD discipline.

● Stay current with advances in GenAI, agentic AI, and MLOps — bringing relevant innovations to

the team.

Required Skills & Experience

Software Engineering

● 3–5 years of software engineering experience, including at least 2 years in ML/AI engineering

roles.

● Strong Python development skills; familiarity with Java or Node.js is a plus.

● Solid understanding of distributed systems and data pipeline design.

● Containerization experience with Docker; basic Kubernetes knowledge.

AI / Machine Learning

● Hands-on experience building and deploying traditional ML models (classification, regression,

clustering, anomaly detection).

● Proficiency with ML frameworks: scikit-learn, PyTorch, or TensorFlow.

● Real-world experience delivering at least 1–2 LLM or GenAI applications into production.

● Familiarity with RAG architectures and vector search.

● Working knowledge of prompt engineering and LLM evaluation techniques.

● Experience with agentic AI frameworks (LangChain, LangGraph, AutoGen, or equivalent).

Cloud & DevOps

● Hands-on experience with at least one major cloud platform: AWS, Azure, or GCP.

● Familiarity with CI/CD pipelines and deployment automation.

● Understanding of model versioning, code versioning, and configuration management.

Data & Databases

● Experience working with SQL databases and NoSQL stores.

● Familiarity with vector databases (ChromaDB, Pinecone, pgvector, or equivalent) for embedding

based search.

● Ability to build and maintain data ingestion and feature engineering pipelines.

Observability & Production Readiness

● Experience implementing logging, monitoring, and alerting for production AI systems.

● Familiarity with resilience patterns: rate limiting, failover, circuit breakers.

Banking & Compliance Context

Banking is a regulated environment. While deep compliance expertise is not required at this level,

you should be:

● Aware of the importance of explainability, fairness, and auditability in AI models used for

financial decisions (credit, fraud, risk scoring).

● Comfortable implementing AI guardrails and safety controls to meet risk, compliance, and audit

requirements.

● Willing to work within and learn the organization's AI governance and responsible AI

frameworks.

● Mindful of data privacy, PII handling, and secure engineering practices — especially under GDPR,

RBI, or equivalent regulatory regimes.

Good to Have

● Prior experience in banking, financial services, fintech, or payments.

● Exposure to AI governance, model risk management, or responsible AI frameworks.

● Experience with graph databases (e.g., Neo4j) for fraud network or knowledge graph use cases.

● Contributions to open-source AI/ML projects or published work in GenAI.

● Experience with MLOps tooling: model monitoring, retraining pipelines, experiment tracking

(MLflow, Weights & Biases).

● Familiarity with multi-agent architectures for complex workflow automation.