// Selected enterprise engagements — strategy, architecture, and delivery leadership
All client identifiers anonymized.
Housing Finance / BFSI
In-flight
AI-Assisted Property Valuation Engine
For a Leading Indian Housing Finance Lender
Led the design of a dual-path valuation engine combining a rules-based baseline with an AI inference layer. Authored the BRD, technical approach, data specs, and governance wrapper — with explainability and human-in-the-loop on high-variance cases.
Rules + LLM Hybrid
3rd-Party Data Integ.
Explainability
HITL Governance
Education / Non-Profit
In-flight
Automated Field-Team Nudging Engine
For a National Education Foundation (India, Multi-State)
Led design & delivery of a context-aware nudging engine for thousands of distributed field members. Orchestration backbone connects a workflow layer, an LLM reasoning layer, and a field-activity document store. Program managers launch and A/B-test nudges without engineering.
n8n Orchestration
LLM Reasoning
Behavior Triggers
Nudge Guardrails
Cloud / Audit
POC Delivered
RAG Guidance Layer for Cloud Partner Audits
For a Global Hyperscaler Audit Firm
Led delivery of a RAG POC grounded in gold-standard historical audit reports — surfacing context-aware recommendations, evidence cues, and narrative scaffolding to auditors in-line. Wrote the dev brief, BRDs, demo scripts, and designed citation-traceability guardrails.
Python Service
Vector Store
Controlled LLM
Citation Guardrails
BFSI / Asset Mgmt
In Refinement
Institutional Equity Research Pipeline
For an APAC BFSI Asset Manager
Turned a failed GenAI research experiment (hallucinations, poor attribution) into a production-grade, sourcing-first pipeline. Decomposed the note lifecycle into grounded stages, added hallucination-detection cross-checks, and enforced a citation threshold on every draft.
Multi-Stage RAG
Structured Financials
Numeric Verification
Citation Discipline
Pharmaceuticals
Delivered
AI Replica Personas for Leadership
For a Global Specialty Pharma Company
Led the design of rigorously crafted AI persona system prompts and scoped knowledge bases for two leadership voices. Authored the methodology (background, decision frameworks, communication patterns, guardrails) so personas stay in-voice and refuse to speak outside competence.
System Prompt Design
Curated KB
Voice Fidelity
Competence Guardrails
Enterprise AI Thought Leadership
Final Editing
Enterprise AI Context Readiness — White Paper
Research-Backed Executive Framework
Co-developed the thesis that enterprise AI fails on context readiness, not model capability. Built a multi-component framework: AI SaaS viability scoring, context maturity modeling, retrieval reliability scoring, and an enterprise AI economics canvas — via a 20+ stage automated research pipeline.
Context Maturity
Retrieval Reliability
AI Economics
SaaS Viability
BFSI / Capital Markets
Delivered
BPO Services Framework & FTE Sizing
For a Singapore-Based Securities House Engagement
Catalogued the prospect's operations into 93 processes across 14 functional categories. Built a bottom-up, defensible FTE estimate anchored in complexity, volume, and automation potential — packaged as a drill-down Excel workbook usable directly in negotiations.
Process Taxonomy
FTE Estimation
Automation Potential
Excel Workbook
Professional Services
Delivered
EOR & Recruitment Partnership Proposal
For an APAC AI-Native Startup Mandate
Rebuilt a commoditized staffing proposal into a strategic co-builder narrative — restructured around hiring archetypes, jurisdictional complexity, and speed-of-hire. Tied commercial model to hiring velocity and country coverage, not per-seat rates.
Reusable
Proposal Pattern
Commercial Model
Hiring Archetypes
Strategic Positioning
Branded Deliverable