Software Engineer III

Bank of America•Charlotte, NC

8d•Onsite

About The Position

This job is responsible for developing and delivering complex requirements to accomplish business goals. Key responsibilities of the job include ensuring that software is developed to meet functional, non-functional and compliance requirements, and solutions are well designed with maintainability/ease of integration and testing built-in from the outset. Job expectations include a strong knowledge of development and testing practices common to the industry and design and architectural patterns. The Software Engineer III role is a hands-on individual contributor position responsible for designing, developing, testing, and supporting enterprise-grade applications and services for the Legal & Public Policy department. The engineer will be a member of the Legal AI development team, which builds AI-enabled solutions that improve how legal professionals access information, analyze documents, summarize complex content, and interact with legal knowledge and technology platforms. The role will contribute to applications that process and summarize legal documents, provide conversational access to legal knowledge, integrate with internal and external Legal AI platforms, and enable reusable AI agents and workflows. The role spans Python services, document ingestion and parsing, APIs, AI orchestration, retrieval pipelines, search technologies, application integrations, and potentially front-end user experiences. The engineer will work in close partnership with Legal stakeholders and technology teams to deliver solutions that are secure, maintainable, explainable, and appropriate for use in a regulated enterprise environment.

Requirements

Strong hands-on experience developing enterprise applications and services using Python.
Experience designing and implementing RESTful APIs using FastAPI, Flask, Django, or comparable Python frameworks. FastAPI experience is strongly preferred.
Strong understanding of object-oriented programming, modular application design, exception handling, dependency management, and reusable software components.
Experience processing PDF files and other unstructured document formats using Python libraries and content extraction frameworks.
Experience with one or more document processing libraries or platforms such as PyMuPDF, pypdf, pdfplumber, Apache Tika, Unstructured, python-docx, or comparable technologies.
Experience integrating applications with internal or external platforms using REST APIs, SDKs, authentication mechanisms, and structured message formats such as JSON.
Experience developing automated unit and integration tests using tools such as pytest or comparable testing frameworks.
Working knowledge of relational or non-relational databases and the ability to design appropriate application data access patterns.
Understanding of secure API and application development practices, including authentication, authorization, input validation, secrets protection, and dependency vulnerability management.
Experience using Git-based source control and working with automated build, test, and deployment pipelines.
Ability to diagnose and resolve complex technical issues across application, API, data, and integration layers.
Strong analytical and problem-solving skills with the ability to evaluate implementation options and make pragmatic engineering decisions.
Strong written and verbal communication skills, including the ability to communicate technical concepts to both engineering and non-engineering stakeholders.
Demonstrated ability to work collaboratively in an Agile product development environment.
Intellectual curiosity and a strong appetite to learn new technologies, frameworks, integration patterns, and AI engineering practices.

Nice To Haves

Experience developing AI or Generative AI applications using enterprise-approved large language models and supporting platforms.
Hands-on experience with AI orchestration frameworks such as LangChain, LangGraph, Semantic Kernel, or comparable frameworks.
Experience building agentic workflows involving tool calling, workflow routing, state management, prompt orchestration, and multi-step reasoning.
Experience implementing or contributing to retrieval augmented generation pipelines, including ingestion, parsing, chunking, metadata enrichment, embeddings, vector indexing, retrieval, reranking, and grounded response generation.
Experience with Elasticsearch, including document indexing, text search, metadata filtering, relevance tuning, and hybrid keyword and semantic search concepts.
Familiarity with vector databases, embedding models, semantic search, and retrieval evaluation techniques.
Familiarity with graph database concepts and technologies such as Neo4j, including the use of entities and relationships to represent interconnected knowledge.
Experience building or consuming integrations based on Model Context Protocol.
Experience developing AI agents, connectors, actions, or workflows using Microsoft Copilot Studio.
Experience integrating with enterprise SaaS platforms or AI vendor systems through APIs, SDKs, webhooks, or MCP interfaces.
Experience with Angular and TypeScript for the development of enterprise web applications and conversational user experiences.
Familiarity with cloud-native or containerized application development using Docker, Kubernetes, or equivalent enterprise platforms.
Familiarity with document OCR, image-based document extraction, table extraction, document classification, and multimodal content processing.
Understanding of AI evaluation concepts, including groundedness, retrieval quality, response relevance, traceability, hallucination assessment, and human feedback.
Familiarity with responsible AI, data privacy, security, records management, and governance considerations for AI-enabled applications in regulated environments.
Experience applying observability practices to distributed applications, including structured logging, metrics, tracing, monitoring, and operational alerting.
Experience working with legal documents, legal technology platforms, knowledge management systems, or other document-intensive business domains.

Responsibilities

Codes solutions and unit test to deliver a requirement/story per the defined acceptance criteria and compliance requirements
Design, develop, test, deploy, and support Python-based applications, reusable services, and APIs that enable Legal AI capabilities
Build RESTful services using Python frameworks such as FastAPI, including request validation, exception handling, authentication and authorization integration, logging, and API documentation
Develop document ingestion and processing capabilities for PDF, Microsoft Office, text, email, and other structured and unstructured content formats
Evaluate and use appropriate Python document processing libraries for text extraction, document parsing, metadata extraction, content normalization, table extraction, and preparation of documents for downstream AI processing
Implement solutions for AI-assisted document summarization, question answering, information extraction, classification, and conversational experiences
Contribute to the development of Ask Legal chatbot capabilities that allow users to search, retrieve, and interact with approved legal knowledge sources
Integrate internal applications with Legal AI vendor platforms using secure Python APIs, SDKs, web services, and Model Context Protocol client and server patterns
Build and maintain AI orchestration workflows using frameworks such as LangChain, LangGraph, or equivalent bank-approved technologies
Develop and integrate tools, functions, prompts, state management, routing logic, and multi-step workflows used by AI agents
Support the enablement of AI agents using Microsoft Copilot Studio and other approved enterprise AI platforms and frameworks
Contribute to retrieval augmented generation pipelines, including document ingestion, parsing, chunking, metadata enrichment, embedding generation, indexing, retrieval, reranking, prompt assembly, and response grounding
Develop search and retrieval capabilities using Elasticsearch, including keyword, metadata, and semantic or vector-based search patterns where appropriate
Work with relational databases, document databases, vector stores, and graph databases based on solution requirements.
Apply secure software engineering practices appropriate for a regulated enterprise, including input validation, access control, secrets management, data protection, audit logging, dependency management, and vulnerability remediation
Develop automated unit, integration, API, and regression tests to improve solution quality and reliability
Participate in code reviews, design discussions, technical troubleshooting, and continuous improvement of team development practices
Collaborate with Legal stakeholders, product managers, architects, platform teams, cybersecurity, risk partners, and other engineering teams to translate business needs into reliable technical solutions.
Investigate production and non-production issues across application, API, integration, data, search, and AI orchestration layers.
Create and maintain technical documentation, including API specifications, implementation designs, support procedures, data flows, and operational runbooks.
Conduct proof of concept and technical spikes to evaluate emerging libraries, frameworks, AI patterns, and integration approaches
Share technical knowledge with team members and contribute reusable components and patterns that accelerate delivery across Legal AI initiatives