Technical Lead — Agentic AI & MVP

Job description

**Job Title:** Senior DevOps Engineer – Tech Lead — Agentic AI & DevOps **Department** : CIB ITO Infrastructure **About Business line/Function:** The Mumbai Innovation Lab operates on two parallel tracks: delivering production-grade AI agents and MVPs for business units across the bank, and conducting pure innovation work — exploring, prototyping, and pushing the boundaries of what agentic AI can do in a regulated financial environment. As Tech Lead, you will be the technical reference for both tracks in Mumbai, working within and contributing to an established agentic framework already deployed across the group. Your scope is intentionally cross-functional (IT operations, compliance, risk, business lines) and cross-regional (Mumbai ↔ Paris and other group entities). You will translate architecture decisions into delivered systems, mentor the Mumbai team, and actively contribute back to the shared framework — improving it, extending it, and demonstrating its value on real use cases. ****Responsibilities**** Direct Responsibilities - Lead technical delivery of AI agents and MVPs for cross-functional business units — IT ops, compliance, risk, business lines — within the group's established agentic framework - Contribute to and extend the shared agentic framework: propose improvements, develop reusable components, document patterns for group-wide adoption - Drive pure innovation work in parallel: prototype emerging approaches (LightRAG, TigerRAG, graph-based RAG, advanced MCP patterns) and evaluate their production potential - Design agentic workflows using LangChain, LangGraph, and LangServe — multi-agent reasoning, tool-calling, stateful orchestration, API/microservice exposure - Build production assistants on real infra datasets: incident investigation, RCA, change analysis, CMDB querying, run assistance, documentation search - Implement and enforce guardrails: hallucination control, access control, secrets management, output validation, sensitive data handling - Own end-to-end observability via LangSmith: prompt tracing, tool call monitoring, evaluation datasets, performance alerting Contributing Responsibilities - Coordinate technically across Paris, Mumbai, and group entities — architecture reviews, cross-regional ceremonies, async collaboration - Provide technical leadership and mentorship to the Mumbai engineering team - Deploy and operate POC environments on cloud-native and containerized platforms (Docker, Kubernetes, Helm, GitOps, ArgoCD, GitLab CI/CD) - Formalize architecture patterns and contribute to industrialization specs for build teams across the group **Technical & Behavioral Competencies** **Technical Skills:** - LightRAG, TigerRAG, or graph-based retrieval experience - Contribute to a live agentic framework used across a CAC 40-tier banking group - MCP server design and connector architecture - Go or TypeScript alongside Python - CI/CD & GIT Ops (Expert) - Observability (Intermediate) **Transversal Skills:** - Knowledge of DPDP Act (India) and/or GDPR - Cross-functional stakeholder engagement (IT ops, risk, compliance, business lines) - Dual mandate: ship MVPs AND do real innovation — not one or the other - Ability to work with Agile methods (Mastered) - Cross-functional and cross-regional scope from day one - Tech Lead path with direct line to Global Head of Innovation ****Specific Qualifications:**** **Skills Referential** **(** **Required knowledge, skills and abilities)** ****Technical Skills:**** - 5+ years in DevOps / AI/ML engineering, with 2+ years on LLM-based agentic systems in production - Proven experience contributing to or extending a shared technical framework — not just consuming it - Strong Linux, distributed systems, networking, and enterprise infrastructure architecture expertise - Mastery of containerized environments: Docker, Kubernetes, Helm, GitOps, GitLab CI/CD, ArgoCD - Hands-on LangChain, LangGraph, LangServe, LangSmith — real production deployments - RAG architecture depth: chunking, embeddings, vector stores, hybrid search, grounding strategies - Solid classical observability: Prometheus, Grafana, ELK/OpenSearch, OpenTelemetry ****Transversal Skills:**** - Strong grasp of AI security: access control, data protection, auditability, guardrails, secrets management - Experience in a regulated industry — banking, insurance — with compliance and governance awareness - Fluent professional English — cross-regional technical leadership and documentation are daily ****Behavioral Skills:**** - Ability to share/transmit knowledge - Be results oriented - Creativity & Innovation / Ability to solve problems - Ability to collaborate / teamwork - Develop and maintain system documentation, including configuration guides, and standard operating procedures. - Direct and be responsible for the implementation effort. - Provide technical guidance and mentorship to team members. - Able to work autonomously and as part of a team using strong analytical skills. - Be service oriented, customer focused, positive, committed and have an enthusiastic “can do” attitude. - Demonstrate a systematic and logical approach to problem-solving. - Able to follow the bank’s standards, processes, and procedures. - Escalating incidents internally or to 3rd party partners when required. **Education Level:** Bachelor’s or Master’s degree in Computer Science, Information Technology or Equivalent **Experience level:** 7-9 +Years production experience as Senior lead under Agent AI & Devops **Location:** Mumbai / Bangalore ****About BNP Paribas Group:**** BNP Paribas is the European Union’s leading bank and key player in international banking. It operates in 65 countries and has nearly 185,000 employees, including more than 145,000 in Europe. The Group has key positions in its three main fields of activity: Commercial, Personal Banking & Services for the Group’s commercial & personal banking and several specialized businesses incl