Role Overview
We are seeking an experienced and visionary Senior Lead Python AI/ML Engineer with Generative AI Capabilities to lead the design, development, and deployment of next-generation artificial intelligence and machine learning solutions based out of our Chennai center (operating in a Hybrid work model). With 10 to 12 years of hands-on software and machine learning engineering experience, the ideal candidate will drive the convergence of traditional predictive ML models and modern Foundation Model / GenAI architectures.
In this role, you will be instrumental in building high-scale, low-latency AI solutions applied to Fraud Mitigation, Risk Scoring, and Loss Prevention across US Retail Banking products
2. Key Responsibilities
AI/ML & Generative AI Architecture
- Architect, build, and deploy production-grade Generative AI systems, including Retrieval-Augmented Generation (RAG), autonomous agents, semantic search pipelines, and multi-modal AI workflows.
- Integrate state-of-the-art Large Language Models (LLMs)—including Google Gemini, Anthropic Claude, and open-source models (Llama, Mistral)—optimizing for latency, cost, and accuracy.
- Build domain-specific prompt engineering frameworks, fine-tuning pipelines (LoRA/QLoRA), model evaluation benchmarks, and guardrails to minimize hallucinations and bias.
- Lead the architecture of predictive ML algorithms (anomaly detection, graph neural networks, supervised/unsupervised classifiers) tailored to real-time risk assessment.
Software Engineering & System Scalability
- Design resilient, low-latency, and horizontally scalable microservices and APIs in Python (FastAPI, Flask, AsyncIO) handling high-throughput transactional traffic.
- Integrate vector databases (e.g., Pinecone, Milvus, Qdrant, Chroma, pgvector) with distributed caching and event streaming platforms (Kafka, RabbitMQ).
- Implement robust MLOps and LLMOps practices (CI/CD pipelines, automated model retraining, observability with MLflow/Weights & Biases, model drift monitoring).
Fraud Analytics & Domain Innovation
- Collaborate closely with US-based and global Risk, Compliance, and Loss Prevention teams to build AI solutions targeting identity theft, account takeover (ATO), synthetic identity fraud, card-not-present (CNP) fraud, and transaction anomaly detection.
- Utilize unstructured data (investigator notes, SAR filings, customer interaction logs) via GenAI to automate fraud triage and investigative summarization.
- Ensure strict adherence to US Banking regulatory standards (FCRA, GLBA, Model Risk Management SR 11-7, Fair Lending, Explainable AI / XAI).
Technical Leadership & Mentorship
- Mentor senior/mid-level data scientists and software engineers across the Chennai engineering hub in modern GenAI patterns, clean code principles, and production readiness.
- Drive technical roadmap, evaluate emerging AI frameworks, conduct proof-of-concepts (PoCs), and present architectural proposals to global engineering leadership.
3. Required Qualifications & Experience
- Education: Bachelor’s or Master’s degree in Computer Science, Data Science, Artificial Intelligence, Applied Mathematics, or a related quantitative field.
- Total Experience: 10 to 12 years of professional experience in software engineering and production AI/ML system development.
- Core Python Expertise: Advanced proficiency in Python 3.x (multiprocessing/asyncio, design patterns, profiling, packaging, and clean code paradigms).
- Core ML Frameworks: Deep practical experience with PyTorch, TensorFlow, Scikit-learn, XGBoost, and LightGBM.
- Generative AI Proficiency:
- Demonstrated success building production GenAI applications (LangChain, LlamaIndex, CrewAI, AutoGen).
- Hands-on experience with embedding models, dense retrieval, rerankers, and vector indexing strategies.
- Distributed Systems & Cloud: Strong hands-on experience with cloud platforms (AWS, GCP, or Azure), containerization (Docker, Kubernetes), and distributed data processing (PySpark, Ray).
- MLOps / LLMOps: Proven track record in orchestrating end-to-end ML lifecycles from data validation to real-time inference monitoring.
4. Preferred & Good-to-Have Skills
- Google Gemini & Anthropic Claude: Hands-on experience with API integration, function calling/tools, context caching, and prompt optimization specifically for Google Gemini (Pro/Flash) and Anthropic Claude (Opus/Sonnet/Haiku).
- Domain Knowledge (Fraud & Loss Prevention):
- Experience designing fraud rule engines, anomaly detection platforms, or predictive risk models.
- Familiarity with transaction monitoring, biometric behavioral analytics, and fraud consortium datasets.
5. Technical Skills Matrix
Category
Technologies & Tools
Programming Languages
Python (Advanced), SQL, Bash
Generative AI & LLMs
Anthropic Claude, Google Gemini, OpenAI GPT, Open-source LLMs (Llama, Mistral), LangChain, LlamaIndex, DSPy
Classical & Deep ML
PyTorch, TensorFlow, Scikit-Learn, XGBoost, LightGBM, Hugging Face Transformers
Vector DBs & Search
Pinecone, Milvus, Qdrant, Weaviate, FAISS, pgvector, Elasticsearch
Backend & APIs
FastAPI, Flask, AsyncIO, gRPC, REST
Data & Streaming
Apache Kafka, Apache Spark / PySpark, Pandas, Polars, DuckDB
MLOps & DevOps
MLflow, Kubeflow, Docker, Kubernetes, GitHub Actions, Terraform, Prometheus/Grafana
Domain Tools & Standards
Model Risk Management (SR 11-7), Explainable AI (SHAP, LIME), PII Masking
6. Key Performance Indicators (KPIs)
- Model Performance & Accuracy: Precision/Recall trade-offs in fraud detection, false positive reduction, and sub-100ms real-time inference latency.
- Engineering Excellence: Code quality, test coverage (>85%), modular design, and robust error handling.
- Business Impact: Measurable reduction in fraud loss basis points and acceleration of investigator workflow efficiency through GenAI summarization/triage.
- Mentorship & Global Delivery: On-time milestone delivery, technical capability elevation across local and global engineering pods.
------------------------------------------------------
Job Family Group: Technology------------------------------------------------------
Job Family:Applications Development------------------------------------------------------
Time Type:Full time------------------------------------------------------
Most Relevant Skills Please see the requirements listed above.------------------------------------------------------
Other Relevant Skills For complementary skills, please see above and/or contact the recruiter.------------------------------------------------------
Citi is an equal opportunity employer, and qualified candidates will receive consideration without regard to their race, color, religion, sex, sexual orientation, gender identity, national origin, disability, status as a protected veteran, or any other characteristic protected by law.
If you are a person with a disability and need a reasonable accommodation to use our search tools and/or apply for a career opportunity review Accessibility at Citi.
View Citi’s EEO Policy Statement and the Know Your Rights poster.
Skills Required
- Bachelor's or Master's degree in Computer Science, Data Science, Artificial Intelligence, Applied Mathematics, or a related quantitative field
- 10 to 12 years of professional experience in software engineering and production AI/ML system development
- Advanced proficiency in Python 3.x, including multiprocessing, asyncio, design patterns, profiling, packaging, and clean code
- Deep practical experience with PyTorch, TensorFlow, Scikit-learn, XGBoost, and LightGBM
- Production experience building generative AI applications using tools such as LangChain, LlamaIndex, CrewAI, or AutoGen
- Hands-on experience with embedding models, dense retrieval, rerankers, and vector indexing strategies
- Strong experience with AWS, GCP, or Azure; Docker; Kubernetes; and distributed processing with PySpark or Ray
- Experience orchestrating end-to-end ML lifecycles from data validation through real-time inference monitoring
- Hands-on experience integrating Google Gemini and Anthropic Claude APIs, including function calling, context caching, and prompt optimization
- Experience designing fraud rule engines, anomaly detection platforms, or predictive risk models
- Familiarity with transaction monitoring, biometric behavioral analytics, and fraud consortium datasets
Citi Compensation & Benefits Highlights
The following summarizes recurring compensation and benefits themes identified from responses generated by popular LLMs to common candidate questions about Citi and has not been reviewed or approved by Citi.
-
Healthcare Strength — Health coverage is presented as comprehensive, including free in-network preventive care, mental-health resources, and the Be Well Program. Materials also state new hires have access to health and insurance benefits from the outset.
-
Retirement Support — The retirement plan offers a $1-for-$1 match up to 6% of eligible pay. New employees are eligible for the plan, with automatic enrollment after 90 days if they do not enroll themselves.
-
Inclusive Benefits Coverage — Benefits highlight fertility treatment coverage, adoption/surrogacy reimbursement up to $30,000, and disability coverage. Coverage extends to spouses and domestic partners regardless of gender identity or sexuality, with gender-affirming care-related support noted.
Citi Insights
What We Do
Citi's mission is to serve as a trusted partner to our clients by responsibly providing financial services that enable growth and economic progress. Our core activities are safeguarding assets, lending money, making payments and accessing the capital markets on behalf of our clients. We have 200 years of experience helping our clients meet the world's toughest challenges and embrace its greatest opportunities. We are Citi, the global bank – an institution connecting millions of people across hundreds of countries and cities.
