Audit Friction
Map processes, uncover hidden bottlenecks, and isolate manual tasks ripe for automation.
I architect AI systems and Salesforce integrations that process 200,000+ cases monthly with >99% accuracy across financial services — delivering 20+ FTE saved, 80%+ adoption across 1,500+ users at a KKR-backed financial infrastructure firm.
20+
FTE Saved Annualized
Through AI systems & automation
200K+
Cases Monthly
>99% routing accuracy
1,500+
Active Users
80%+ organic adoption rate
55%
Faster Onboarding
KYC cut from 18 to 8 days
71
AI Initiatives
Delivered via DigiOps Programme
Methodology
A rigorous process designed to turn operational friction into measurable efficiency.
Map processes, uncover hidden bottlenecks, and isolate manual tasks ripe for automation.
Structure secure RAG loops, agentic workflows, and human-in-the-loop validation checkpoints.
Develop pipelines with Python/SQL, call Vertex AI models, and bind systems using robust APIs.
Build intuitive UI surfaces inside existing platforms (e.g. Salesforce) to push organic adoption.
Validate FTE reductions, enforce strict compliance auditing, and tune model latency metrics.
"Here is how this methodology played out across three enterprise transformations at a KKR-backed financial infrastructure firm."
Portfolio
Manual tracking across 300+ employees
Skill gaps and workload spread tracked in disconnected spreadsheets with no executive visibility.
Role-Based Self-Service Platform
Real-time BAU vs project visibility, bulk manager overrides, and executive telemetry dashboards.
16% info-request case volume
Support agents scanning manuals manually. High escalation rate.
Instant AI Answers
3 RAG bots in Salesforce with cited manuals, cutting escalation to 3% across 24 knowledge initiatives.
Part of the team that designed and rolled out three RAG-based AI assistants powered by Vertex AI.
18-day manual vetting
Compliance officers vetting checklists and reference data offline.
8-day automated cycle
AI checklist extractor, 751 monthly screenings, real-time registry screening.
Manual L1 triage burden
Inbound cases manually screened, routed, and categorized causing delays.
Autonomous Triage at Scale
Background routing, mass noise deletion, and duplicate flagging across 215,570+ monthly cases.
+1.88 FTE from duplicate flagging (15,206 monthly checks)
Reactive escalation handling
Reactive escalation handling — negative sentiment and high-risk tickets identified only after client impact, leading to delayed intervention
Predictive Intervention
Predictive Intervention — AI prompts I designed classify 20,000+ emails across MarkitWire and TriResolve with dual-category scoring (Client Services + Product) enabling proactive outreach before escalation
Designed all AI prompts for sentiment classification and led change management across Client Services teams. The dual-category framework (positive, negative, neutral) for both Client Services and Product domains enables managers to intervene proactively before clients escalate. Supported team-built integration of live call transcription and Salesforce summary glances.
Manual alert triage
Complex alerts including LimitHub and TOC FIX disconnections required reactive human investigation.
Autonomous Resolution
Automated investigation, commentary, and routing of non-actionable alerts without human initiation.
Manual trade pipeline mapping
Fragmented data syncs between Service Cloud and MarkitWire, creating operation blindspots and manual tracking bottlenecks.
Unified REST Connectors
Real-time Salesforce-MarkitWire REST connectors, Apex handler queues, and automated SOX permission validations.
Scale
The systems I architect and contribute to operate in production across global financial services workflows.
200K+
Cases routed monthly
Auto-Triaging Engine
Personal Lead20K+
Client emails analyzed
Sentiment Intelligence (team)
Team Contribution4,500+
System alerts handled monthly
Failure Management
Personal Lead751
Compliance screenings monthly
KYC Automation
Personal Lead71
AI initiatives delivered
DigiOps Programme
Programme Contribution1,500+
Active enterprise users
Cross-Programme Adoption
Personal LeadData Intelligence
Every AI product I ship includes a reporting layer that answers the questions leadership actually cares about.
Key Question How fast can we route 200,000+ cases monthly?
>99% Accuracy
Key Question Which clients are at risk before they escalate?
20K+ Emails Analyzed
Key Question Are all 751 monthly screenings compliant?
100% Compliance
Key Question Who has capacity and what skills do we need for cross-training?
300+ Employees
Key Question Which 4,500 alerts need human attention?
Autonomous Routing
Key Question Are our AI initiatives on track?
71 Initiatives Delivered
Enterprise Stack
Interactive breakdown of my transformation stack, mapped directly to strategic decision systems.
Transformation Stack Layers
Mindset
Every transformation I have shipped started with one question. Most operational friction survives because nobody asks it.
"Why do we still do it this way?"
That single question has driven every product I have built. Most processes survive not because they work, but because nobody challenges them.
NEXUS Platform — 300+ employees, real-time skill & ops telemetry
Dot & ODA RAG Assistants — 70%+ self-service resolution
Intelligent Onboarding — 55% cycle time reduction
Curiosity is not a personality trait. It is a professional method.
Authority
Vetted achievements, media features, and internal honors from leading institutions.
Recognized among leading young operators scaling enterprise AI and automation systems. Featured for designing advanced intelligent operations pipelines and RAG workflows that deliver quantifiable business transformation.
Awarded for delivering Looker analytics and Salesforce auto-triaging models that reduced manual queue noise for global operations.
Awarded for building a serverless SOX compliance engine that replaced manual user access checks, saving 4.1 FTE.
Won first place for architecting a user-centered product design solution to optimize high-impact operational friction.
Received multiple quarterly spot awards for rapid AI tooling deployment and continuous process improvement contributions.
Timeline
A track record of driving process automation, mobile deployment, and operational design.
OSTTRA (KKR Portfolio). Architected AI systems and Salesforce integrations processing 200,000+ monthly cases with >99% accuracy. Delivered 20+ FTE in total annualized savings and 80%+ adoption across 1,500+ users.
OSTTRA. Engineered SOX Compliance automation using Google Apps Script, eliminating 4.1 FTEs of manual audit work and compressing cycles from 45 to 5 days. Unlocked 2+ FTEs across 10+ Lean Six Sigma initiatives. Directed transition from Microsoft to Google Workspace, freeing 0.4 FTEs.
Two consecutive React Native developer roles at RPQ IT Services (Noida, Sector 16, Apr 2021–May 2022) and Markup Designer (Noida, Sector 62, Nov 2020–Mar 2021). Managed full SDLC for iOS and Android, UI/UX in Figma, and CI/CD pipelines.
Dual MBA + PGPM in Finance & Analytics. Lean Six Sigma certification. Focused on process optimization, financial analytics, and structured problem-solving methods.
ICFAI Business School, Gurgaon · CGPA: 9.0/10
IFHE Hyderabad (Online) · CGPA: 8.75/10
Birla Institute of Technology, Mesra · CGPA: 7.52/10
Authenticity
Brief log of lessons, experiments, and learnings as a systems builder.
"Accidentally broke a production sheet during a 2024 database migration. Reinforced why read-only API scopes are vital."
"Discovering that prompting is a new layer of code abstraction. Building a custom Gemini CLI changed my dev pipeline."
"Deep Work by Cal Newport. Applying structured focus methods to design clean, complex RAG assistant pipelines."
"If a manual operational process takes more than 10 minutes, I will spend 10 hours building a clean automation loop for it."
"Building AI is 20% of the work. Getting 1,500 people to use it is the other 80%."
Consultation & Engage
I am open to strategic consultations on intelligent operations, process auditing, and scaling custom AI automation systems.