Why Three Cohorts, Not One Curriculum
A 0-2 year professional and a 10+ year architect walking into the same generic "AI training" both walk out disappointed, for opposite reasons: one is drowning in jargon with nowhere to apply it, the other is bored reviewing basics they mastered a decade ago in a different discipline. Same underlying content library — the entire AI Career Compass material built across this program — re-sequenced and re-weighted for what each career stage actually needs.
About VESPA Analytics & Consulting
AI Career Compass is delivered under VESPA Analytics & Consulting, co-founded by senior industry professionals in Bengaluru in 2014, specifically to bridge the gap between industry and academia — where most AI training comes from one side or the other, rarely both. VESPA operates four connected practice lines — Consulting, Mentoring, CoE Building, and the AI Practitioner Masterclass — and this entire tiered training architecture is how the Mentoring and Masterclass lines are delivered to individuals and cohorts, rather than to a single enterprise client.
The VESPA Framework
Four practice lines, each a different depth of engagement with the same underlying expertise — and each maps directly onto a tab in this document.
Consulting
Short, diagnostic engagements — scoping a problem before committing to a build.
₹15,000–35,000 / half-dayMentoring
45 hours over 9 weekends, with pre-course and post-course materials and mentoring — this is all three cohort tabs (Foundation, Upskilling, and Leadership) in this document.
₹75,000–1,50,000 / seatCoE Building
A 30/60/90-day Build → Enable → Transfer & Scale plan for organizations, not individuals.
₹2,00,000–4,00,000AI Practitioner Masterclass
10 hours across 2 half-days, domain-specific — the only track with its own separate, shorter timeline. This is the standalone Masterclass tab in this document.
Priced on the consulting rate cardWhy This Matters to a Cohort Member
Every mentor in this program also consults and builds CoE capability for real organizations — the same week. What a cohort learns on Saturday is very often the same judgment being applied in a live client engagement that Monday, not a separately maintained teaching syllabus.
Why VESPA Mentoring Is a Different Kind of Mentoring
Taught From Deployed Work, Not Borrowed Cases
The Warranty Analytics platform, HNI Realty Forum, and Optivis/EREI aren't hypothetical — they're systems VESPA scoped, built, and shipped.
One Mentor, Cross-Industry Pattern Recognition
20+ years across manufacturing, automotive, pharma, healthcare, hi-tech, and telecom means a case from one industry regularly reframes a problem in another.
Senior-Led, Not Delegated
Mentoring sessions are delivered by the same senior co-founders who run VESPA's consulting and CoE engagements — not handed to a junior trainer once the sale is made.
A Real Path Beyond the Cohort
A mentee or their employer can move directly from Mentoring into Consulting or CoE Building on the same VESPA rate card — the relationship doesn't end at graduation.
The Three Tiers, Side by Side
| Dimension | 0–2 Years | 4–6 Years | 10+ Years |
|---|---|---|---|
| Mode | Learning / Skilling | Upskilling / Relearning | Unlearning / Reskilling |
| Primary Focus | Context, case studies, capstone-driven experience | Domain & role-based skills, case studies, conceptual clarity, project management, architecture | Architecture, product development, deployment/rollout, support functions |
| Code Emphasis | High — hands-on throughout | Moderate — enough to direct and review, not necessarily write daily | Low by design — reads architecture diagrams and reviews outputs, rarely writes code |
| Destination | A working, demoable capstone | A domain-specific AI initiative they can own end to end | A deployment-ready product or organizational rollout plan |
The Cohort Model (Applies to All Three Tiers)
Delivery Model, All Tiers
| Element | Standard |
|---|---|
| Timeline | 45 hours over 9 weekends for all three cohorts — Foundation, Upskilling, and Leadership. The standalone Masterclass is the one exception, at 10 hours over 2 half-days. |
| Pre-Course & Post-Course | Pre-course material and post-course mentoring are provided for all three cohorts, not just the 9 live weekends themselves |
| Format | Live, cohort-based sessions (on-site or hybrid), following the CCCC method every session |
| Materials | Tabbed HTML session material per week/module, matching the format already used across this program |
| Between-Session Work | A scoped practice task or capstone milestone due before the next session, reviewed live |
| Assessment | Diagnostic only, in the spirit of the New Batch Assessment — used to pace mentoring, never to gate advancement |
CCCC, Re-Weighted Per Tier
The four-step method never changes. What changes is how much time each step gets.
0–2 Years: Concept 20% · Case Study 25% · Code/Practice 30% · Capstone 25%
Heaviest on hands-on practice and the capstone destination — this tier learns by building, and needs the most repetitions.
4–6 Years: Concept 30% · Case Study 30% · Code/Practice 15% · Capstone 25%
Concept and case study rise together — this tier already codes; what it needs is domain judgment and architectural framing.
10+ Years: Concept 35% · Case Study 35% · Code/Practice 5% · Capstone 25%
Code nearly disappears on purpose. Concept and case study dominate, and the "capstone" becomes a rollout or governance plan, not a script.
0–2 Years: The Foundation Cohort
Mode: Learning / Skilling. This tier hasn't built much yet — the job is repetition, real context, and a capstone that becomes their first real portfolio piece. This is the existing AI Career Compass 9-weekend curriculum, largely as already built.
Concept
Case Study
Code/Practice
Capstone
Course Outline
Module 1: Foundations & Landscape
Weeks 1–2The AI/ML/GenAI/Agentic AI spectrum, Python and Git from zero, real cross-industry case studies.
Draws FromModule 2: Statistics & ML Foundations
Weeks 3–4Descriptive to predictive analytics, ANOVA/chi-square, regression, classification, clustering, cross-validation.
Draws FromModule 3: GenAI, RAG & Agents (Applied)
Weeks 5–7LLM basics, prompting, and a first working agent/RAG build — kept practical, not architectural.
Draws FromModule 4: Capstone Completion & Career Launch
Weeks 8–9Finalize, deploy, and demo a working capstone; build the job-prep toolkit around it.
Draws FromOutcomes: How Success Is Measured
Delivery Specifics for This Tier
| Element | This Tier's Version |
|---|---|
| Cadence | 45 hours over 9 weekends, matching the existing program calendar, with pre-course material and post-course mentoring included |
| Practice Load | Highest of the three tiers — a hands-on exercise every session |
| Capstone Support | Weekly milestone check-ins, using the exact worksheets already built into Weeks 3–5 |
| VESPA Mentoring | Delivered under VESPA Analytics & Consulting's Mentoring practice line — the same founder-led, cross-industry mentor as every other tier, not a delegated junior instructor |
4–6 Years: The Upskilling Cohort
Mode: Upskilling / Relearning. This tier already writes code and owns delivery — what's missing is AI-specific domain judgment, architectural framing, and enough project-management fluency to lead an AI initiative, not just contribute to one.
Concept
Case Study
Code/Practice
Capstone
Course Outline
Module 1: Domain & Role-Based AI Judgment
Concept + Case StudyThe five AI job roles mapped to real case studies, and a framework for translating a business problem into an engineering one.
Draws FromModule 2: ML & GenAI Conceptual Depth
Concept + CodeDeeper coverage than the Foundation tier: supervised/unsupervised distinctions, time series, deep learning basics, token economics — enough to direct and review work, not necessarily write it daily.
Draws FromModule 3: Project Management for AI Initiatives
New ContentScoping an AI project honestly — timelines that account for data-quality discovery, stakeholder alignment on what "done" means, and managing a mixed team of engineers and domain experts.
Draws FromModule 4: Architecture & Governance Case Clinics
Mentoring SessionsLive case clinics on real failures — Zillow's drift, Amazon's bias, Air Canada's chatbot ruling, the Replit incident — each dissected for the governance check that was missing.
Draws FromOutcomes: How Success Is Measured
Delivery Specifics for This Tier
| Element | This Tier's Version |
|---|---|
| Cadence | 45 hours over 9 weekends, the same calendar as the Foundation Cohort, with pre-course material and post-course mentoring included |
| Mentoring | 1:1 mentoring session mid-cohort, focused specifically on their own team's live AI initiative |
| Capstone | A domain-specific initiative proposal they could realistically pitch at work, not a teaching dataset exercise |
| VESPA Differentiator | The 1:1 session is delivered by a mentor who is, in the same week, running live VESPA Consulting and CoE Building engagements — feedback comes from current practice, not a static playbook |
10+ Years: The Leadership & Architecture Cohort
Mode: Unlearning / Reskilling. This tier's challenge is rarely technical aptitude — it's unlearning the assumption that AI adoption is primarily a coding problem. The focus shifts to architecture, product development, deployment and rollout, and the support functions that make AI sustainable inside a real organization.
Concept
Case Study
Code/Practice
Capstone
Course Outline — Anchored to the AI Practitioner Masterclass
Module 1: Composite Architecture, Composable by Design
Concept + Real SystemThe Warranty Analytics & Product Pedigree platform's three-stage architecture — deterministic rules, statistical methods, ML ensembles — as a template for staging any AI capability responsibly.
Draws FromModule 2: Product Development for Agentic Systems
Concept + Real SystemTwo real, currently-building agentic products — a PropTech marketplace and an MSME order-management platform — as the material for how to scope, architect, and narrate an agentic product to non-technical buyers.
Draws FromModule 3: Deployment, Rollout & Support Functions
GovernanceWhat changes operationally once a model ships — drift monitoring, rollback discipline, and the specific organizational support functions (not engineering roles) an AI system needs to stay trustworthy in production.
Draws FromModule 4: Your Organization's AI Rollout Plan
CapstoneThe capstone for this tier is never code — it's a governed rollout plan for one real AI initiative in their own organization, presented and defended to the cohort.
Draws FromOutcomes: How Success Is Measured
Delivery Specifics for This Tier
| Element | This Tier's Version |
|---|---|
| Cadence | 45 hours over 9 weekends, the same calendar as the other two cohorts, with pre-course material and post-course mentoring included |
| Format | The same 4-module content anchored to the AI Practitioner Masterclass, paced across a full 9-weekend cohort rather than compressed into 2 days |
| Code Component | Reading architecture diagrams and reviewing agent outputs only — no hands-on scripting expected or required |
| Capstone | A rollout/governance plan, presented to peers, not a demoable application |
| Pricing Basis | Same fixed-offer basis as VESPA's standard Mentoring practice line |
| VESPA Framework Line | This tier sits on VESPA's Mentoring practice line, drawing its content from the Masterclass line — graduates are the natural pipeline into a direct VESPA CoE Building engagement for their organization |
The Honest Design Principle Behind This Tier
Nobody at 10+ years needs to be taught to code from a tutorial. What they need is exposure to real, deployed architecture decisions — including the ones that were wrong at first — so their next AI decision is made with better judgment, not more syntax.
Masterclass: Advanced & Leadership AI Functions
A domain-specific track above the three cohorts — for leaders who don't need to be convinced AI matters, they need the five functions that decide whether their organization's AI investment survives contact with regulation, economics, and scale.
Concept
Case Study
Code/Practice
Capstone
Five Focus Areas, 10 Hours
Gamification in AI Adoption
Using game mechanics — progress, scoring, milestones — to drive genuine internal AI adoption, not just training completion rates.
AI Product Management
Scoping, pricing, and narrating an AI product so non-technical buyers say yes — "sell business solutions, not technology."
AI Governance & Regulatory
Emerging regulation (including the EU AI Act's risk-tiered approach), internal governance boards, and audit-ready documentation.
Economics of AI
ROI modeling, build-vs-buy, and the Fast Money vs. Slow Money framework for sequencing which AI capability gets funded first.
AI Agents & GenAI Strategy
Where agentic AI and MCP-style connection layers create real leverage versus where they add risk without proportionate return.
Course Outline
Module 1: Gamification & Adoption Design
2 hrs · Concept + Case StudyWhy adoption, not model accuracy, is the most common reason enterprise AI initiatives stall — and how gamified milestones (the same principle behind this program's own cohort-and-capstone structure) fix it.
Draws FromModule 2: AI Product Management
2 hrs · Concept + Real SystemScoping an agentic product for a non-technical buyer, using the Observe → Predict → Optimize narrative from a real, currently-building MSME platform.
Draws FromModule 3: AI Governance & Regulatory Landscape
2 hrs · Concept + Case StudyRisk-tiered regulation, internal governance boards, and the exact case studies where a missing governance layer — not a bad model — caused the failure.
Draws FromModule 4: Economics of AI
2 hrs · Concept + FrameworkToken-level cost economics, ROI modeling for an AI initiative, and the Fast Money vs. Slow Money framework used to sequence real funding decisions.
Draws FromModule 5: AI Agents, GenAI & Capstone Synthesis
2 hrs · Case Study + CapstoneA strategic (not technical) view of agentic AI and MCP-style connection layers, closing with each leader's own AI economics-and-governance plan.
Draws FromOutcomes: How Success Is Measured
Delivery Specifics
| Element | Standard |
|---|---|
| Duration | 10 hours, delivered as 2 half-days, with domain-specific focus for advanced and leadership AI functions |
| Audience | Senior leaders, product heads, governance/compliance leads, and strategy roles — not an engineering audience |
| Code Component | Near zero by design — the same principle as the 10+ Leadership Cohort, held even more strictly here |
| Capstone | A one-page AI economics-and-governance plan for one real initiative, presented and challenged by peers |
| Pricing Basis | Priced above the standard AI Practitioner Masterclass rate, reflecting the added regulatory and economics specialization |
| VESPA Framework Line | The most senior expression of VESPA's Masterclass practice line — delivered by the same senior co-founders across all four VESPA practice lines, not a separate faculty |
Why This Sits Above the Three Cohorts, Not Inside Them
Gamification, product economics, and regulatory governance aren't "more advanced ML" — they're a different discipline entirely, closer to strategy and policy than engineering. This masterclass is deliberately its own track, open to graduates of the 10+ Leadership Cohort or equivalent standing.