Senior Process Manager

Job Summary

We are seeking an astute, forward-thinking Senior Business Analyst specializing in Generative AI (GenAI BA) with 6+ years of comprehensive business analysis experience to drive our enterprise cognitive solutions roadmap. In this high-visibility role, you will act as the pivotal bridge translating ambiguous executive business goals into actionable, production-ready AI capabilities.

You will partner closely with business leaders, domain Subject Matter Experts (SMEs), and technical AI engineering teams (GenAI Architects, Data Scientists, and Full-Stack ML Engineers) to discover high-value use cases, formulate detailed functional requirements for Retrieval-Augmented Generation (RAG) applications, define prompt strategies, establish benchmark evaluation rubrics, and orchestrate seamless User Acceptance Testing (UAT).

Key Responsibilities

C. Knowledge Base & Prompt Strategy Scoping

• Knowledge Base Curation: Collaborate with business SMEs and data engineers to identify, catalog, and curate internal enterprise knowledge bases, documents, and taxonomies for RAG indexing.
• RAG Data Structuring: Specify document chunking logic, metadata filtering schemas, and taxonomy tags to maximize semantic vector search retrieval accuracy.
• Prompt Specification: Define foundational prompt requirements: system instructions, tone-of-voice guidelines, persona parameters, few-shot examples, and output formatting templates.
• Guardrails & Ethics: Identify and document guardrail rules for PII/PHI redaction, copyright protection, brand safety, and hallucination containment.

D. AI Evaluation, UAT & Quality Benchmarking

• Ground-Truth Dataset Design: Construct comprehensive ground-truth test datasets containing representative real-world user queries, multi-turn dialogues, and complex edge cases.
• Evaluation Frameworks: Establish objective evaluation rubrics across core GenAI metrics: answer faithfulness, contextual relevance, answer completeness, tone alignment, and toxicity.
• UAT Execution: Orchestrate User Acceptance Testing (UAT) with business stakeholders; track, triage, and analyze model errors, hallucinations, and user feedback.
• Iterative Refinement: Conduct ongoing prompt iterations and parameter tuning alongside prompt engineers to continually refine response fidelity.

E. Stakeholder Enablement & Change Management

• Cross-Functional Liaison: Serve as the primary GenAI evangelist, bridging the communication divide between non-technical stakeholders and deep technical AI engineers.
• User Enablement & Training: Design training materials, playbooks, and prompt guides to accelerate internal end-user adoption and minimize prompt friction.
• Post-Launch Analytics: Monitor post-deployment analytics (user engagement, thumbs up/down feedback, deflection rates) to propose backlog enhancements.

A. AI Use Case Discovery & Feasibility Analysis

• Opportunity Identification: Lead discovery workshops with enterprise stakeholders to uncover workflow bottlenecks and evaluate potential GenAI automation opportunities.
• Feasibility & Solution Fit: Perform structured feasibility assessments comparing Generative AI / Foundation Models against traditional deterministic automation, rules engines, or predictive ML.
• ROI & Business Case Modeling: Develop comprehensive business cases, quantifying business impact, cost-benefit analyses, token expenditure estimates, and projected Return on Investment (ROI).
• Roadmap & Prioritization: Formulate Minimum Viable Product (MVP) roadmaps, prioritizing use cases based on business value, technical readiness, and risk tolerance.

B. Requirements Engineering for Probabilistic & AI Systems

• Functional Documentation: Author robust Business Requirement Documents (BRDs), Functional Specification Documents (FSDs), and Process Flow Diagrams (BPMN) for AI applications.
• Agile User Stories: Draft granular Agile User Stories with explicit 'Given-When-Then' acceptance criteria specifically tailored to probabilistic AI behavior.
• Edge Cases & Fallbacks: Define system boundary specifications: intent classification thresholds, fallback mechanisms, out-of-domain query handling, and human-in-the-loop (HITL) escalation triggers.
• NFR Specification: Document Non-Functional Requirements (NFRs) including response latency thresholds, token rate limits, concurrency targets, and data residency constraints.

Skill Requirements

• Total Experience: Minimum 6+ years of professional IT Business Analysis experience, with at least 2+ years dedicated to AI/ML, NLP, Conversational AI, or GenAI products.
• Education: Bachelor’s degree in Computer Science, Information Systems, Business Administration, Data Analytics, or related discipline.
• GenAI Domain Literacy: Solid conceptual grasp of Foundation Models (GPT-4, Claude, Gemini, Llama), RAG architectures, vector embeddings, semantic search, and agentic workflows.
• Agile & BA Tooling: Proficiency in Agile/Scrum delivery methodologies; expert administrator in Jira, Confluence, and process modeling tools (Visio, Miro, Lucidchart).
• Data & Querying Skills: Working proficiency with SQL for querying relational databases, validating test datasets, and analyzing feature performance.
• Experience Design: Experience defining user personas, customer journey maps, wireframes, and conversation dialogue trees for chatbots / AI assistants.

Communication: Outstanding executive presence, presentation prowess, and technical writing skills with a knack for simplifying complex AI concepts

Other Requirements

4. Preferred Certifications & Domain Knowledge

• Business Analysis Credentials: Certified Business Analysis Professional (CBAP), PMI Professional in Business Analysis (PMI-PBA), or Certified Scrum Product Owner (CSPO).
• Cloud & AI Credentials: AWS Certified AI Practitioner, Microsoft Certified: Azure AI Fundamentals (AI-900), or Azure AI Engineer Associate.
• Industry Verticals: Domain experience in highly regulated verticals: Banking & Financial Services, Insurance, Healthcare, Life Sciences, or Legal Tech.
• Regulatory Frameworks: Familiarity with enterprise AI governance frameworks (EU AI Act, NIST AI RMF, Responsible AI guidelines).

5. Key Performance Indicators (KPIs) & Success Metrics

The performance and effectiveness of the GenAI Business Analyst will be evaluated on the following key metrics:

• 1. User Adoption: Active User Adoption Rate: Driving ≥80% sustained end-user adoption and engagement for newly deployed GenAI capabilities.
• 2. Requirement Precision: Requirements Quality & Sprint Velocity: Maintaining requirement churn and sprint rework below 10% through precise acceptance criteria.
• 3. Delivery Velocity: Time-to-Value Delivery: Accelerating use case progression from initial discovery through PoC and pilot deployment within 6–8 weeks.
• 4. Quality & Faithfulness: AI Accuracy & Deflection Benchmark: Achieving ≥90% response faithfulness and acceptable UAT sign-off prior to production cutover.
• 5. Measurable ROI: Business Value Realization: Demonstrating quantifiable time savings, cost reductions, or revenue lift documented in post-launch reviews.

6. Candidate Evaluation Matrix (For Hiring Panels)

Interview panels should evaluate candidates across the following five core capability dimensions:

Evaluation Area

Competency Focus

Key Interview Signals

GenAI Problem Framing & Feasibility

Distinguishing high-ROI GenAI use cases from non-AI automation

Demonstrates clear logic on when to use RAG vs. fine-tuning vs. deterministic scripts; calculates token cost implications and business feasibility.

AI Requirements & Edge Case Definition

Handling non-deterministic outputs, fallbacks, and user stories

Articulates how to write acceptance criteria for probabilistic outputs; details clear escalation paths for low-confidence answers and hallucination traps.

RAG Knowledge Structuring & Prompting

Curation of unstructured data, metadata tags, and prompt specs

Explains how chunk size, document hierarchy, and metadata schemas impact retrieval accuracy; writes clear few-shot system prompts.

Testing, UAT & Evaluation Rubrics

Benchmarking faithfulness, relevance, and ground-truth validation

Details methodology for curating golden test datasets; explains how to grade AI outputs objectively with business SMEs during UAT.

Stakeholder Translation & Influence

De-hyping GenAI, managing expectations, and change enablement

Proven ability to bridge executive hype with technical reality; shares examples of overcoming end-user skepticism and driving adoption.