Senior Engineer

1 day ago

Chennai Pune, Tamil Nadu, India Money Forward India Full-time
About Money Forward
Driven by our core mission, "Money Forward. Move your life forward," Money Forward is redefining how individuals and businesses interact with financial services through high-impact technology. In partnership with Sumitomo Mitsui Financial Group (SMFG) and Sumitomo Mitsui Banking Corporation (SMBC), we are launching a next-generation digital bank built from the ground up to empower small and medium-sized businesses (SMBs) across Japan.
Operating as an agile fintech startup within our modern tech hubs in Chennai and Pune, our engineering team is creating high-availability, cloud-native financial solutions using modern architectures, AI-driven automation, and slick user-centric web applications.

Role Overview
We are seeking an exceptional Senior Engineer / Tech Lead to spearhead the development of our next-generation AI Agent Platform for our banking ecosystem. In this role, you will design, build, and scale the core infrastructure that enables autonomous LLM-driven agents to execute complex financial workflows, interact with internal systems, and reason across sensitive datasets.
Because we operate in a highly regulated environment, this role requires a dual focus on cutting-edge AI capabilities and uncompromising security, governance, and compliance. You will guide a team of engineers, make foundational architectural decisions, and ensure our AI agents operate safely, predictably, and transparently at enterprise scale.

Key Responsibilities

Agent Engineering & Orchestration

  • Advanced Agentic Patterns: Design and implement sophisticated agent reasoning frameworks (e.g., ReAct, Plan-and-Execute, Reflection, Sub-tasking) to handle multi-step financial inquiries and complex problem-solving.
  • Secure Tool Calling & Integration: Build robust, deterministic function-calling capabilities that allow agents to securely interact with core banking APIs, CRM systems, and external financial data providers.
  • Multi-Agent Systems: Architect orchestration layers where specialized, role-playing agents (e.g., a compliance-checker agent, a data-retrieval agent, and a client-facing agent) collaborate, cross-check work, and hand off tasks to complete complex enterprise workflows.
  • State & Memory Management: Develop scalable architectures for agent memory (short-term working memory, long-term episodic/semantic storage) using vector databases, enabling agents to maintain accurate context across extended user sessions.
  • Agent Simulation & Sandboxing: Create simulation frameworks to rigorously test autonomous agent behaviors, evaluate decision-making accuracy, and identify edge-case failure modes in a safe environment prior to production deployment.

Security, Governance & Risk Management

  • AI Guardrails & Safety: Design and implement robust guardrail systems to ensure LLM outputs are deterministic, factual, and strictly aligned with banking compliance standards.
  • Data Privacy & DLP: Architect Data Loss Prevention (DLP) mechanisms to redact and protect Personally Identifiable Information (PII) and Non-Public Personal Information (NPI) before data interacts with internal or external LLMs.
  • Auditability & Explainability: Build comprehensive telemetry, tracing, and immutable audit logs for all agent actions and reasoning steps to satisfy internal model risk management (MRM) and external regulatory audits.
  • Access Control: Implement fine-grained, Zero-Trust Role-Based Access Control (RBAC) for agent execution, ensuring agents only have access to the data and APIs required for their specific context.

Architecture & Development

  • Secure Platform Design: Design scalable, event-driven architectures capable of managing asynchronous agent workflows, deployed within secure Virtual Private Clouds (VPCs) or on-premise environments.
  • Backend Engineering: Write clean, maintainable, and high-performance code (primarily in Python and/or Java/Go) to support distributed, highly available agent execution.

Technical Leadership

  • Team Mentorship: Lead, mentor, and grow a team of backend and machine learning engineers, instilling a culture of "security by design" and technical excellence.
  • Code & System Review: Conduct rigorous code reviews, threat modeling, and system design evaluations to ensure resilience against prompt injection, data exfiltration, and other LLM vulnerabilities (e.g., OWASP Top 10 for LLMs).
  • Strategic Roadmapping: Partner with Product, Information Security (InfoSec), and Compliance teams to translate business requirements and AI research into a secure, executable engineering roadmap.

Requirements

  • Experience: 7+ years of overall software engineering experience, with at least 2+ years in a technical leadership or team lead capacity.
  • Agent & LLM Expertise: Deep, hands-on experience building production applications using agentic frameworks (e.g., LangGraph, AutoGen, CrewAI, LlamaIndex), implementing semantic search/RAG, and managing LLM context windows efficiently.
  • Backend Expertise: Deep proficiency in Python, Java or Go with a strong track record of building complex, distributed backend systems and microservices in enterprise environments.
  • Security & Compliance: Strong working knowledge of data security principles, encryption (at rest and in transit), IAM, and regulatory frameworks applicable to banking (e.g., SOC 2, PCI-DSS, GLBA).
  • Cloud & Infrastructure: Extensive experience with public cloud platforms (AWS, GCP, or Azure), deploying services in highly restricted network topologies, and container orchestration (Kubernetes).
  • Education: Bachelor’s or Master’s degree in Computer Science, Engineering, or a related field (or equivalent practical experience).
Nice-to-Haves
  • Previous experience working in FinTech, traditional banking, or other highly regulated industries.
  • Familiarity with fine-tuning open-source LLMs (Llama 3, Mistral) and deploying them securely via vLLM or TGI within isolated environments.
  • Experience with Model Risk Management (MRM) guidelines (e.g., SR 11-7) and automated LLM evaluation frameworks (LLM-as-a-judge).
  • Certifications in cloud security or architecture (e.g., AWS Certified Security, CISSP).