01 / MODELLIVING MODEL · ILLUSTRATIVE INPUTS
  1. +0.10ssignals · price, volume, cost, revenue, forecast
  2. +1.80srelationships · 0 resolved
  3. +3.90smodules · drivers / statements / planning
  4. +5.40slayers · finance → data → automation → systems → intelligence
  5. +7.00sstate · adaptive · variance now feeds the forecast
MODEL RESOLVED

I turn financial complexity into intelligent, automated systems.

Muhammad Ahmed Saeed

FP&A Expert at Sunny's · Founder of SpotMAAT · ACCA · BIDA® · FPAP®
Explore the model View case files

An animated model: price, volume and cost combine into revenue, margin and cash; forecast and actual produce variance. The variables organise into three modules — drivers, statements and planning — which sit on five capability layers: finance, data, automation, systems and intelligence. Variance then feeds back into the forecast.

Hover, tap or tab to any node to inspect it.
02 / EXPERIENCETHESIS → TRAJECTORY → IDENTITY
2.1 · THESIS

A finance function is a system. I build it as one.

The model, its data, the pipeline that refreshes it and the report leadership reads are one piece of work, and I build all of it. That means three-statement and forecasting models, Power BI semantic models, and ERP pipelines that produce the recurring reports without anyone rebuilding them by hand.

financedecision supportdataautomationsystems
2.2 · TRAJECTORY
CAPABILITYwhat I can build
  1. 01Accounting
  2. 02FP&A
  3. 03BI
  4. 04Automation
  5. 05Systems
  6. 06AI
ALTITUDEwhere the work sits
  1. 01Transactional finance
  2. 02Management reporting
  3. 03FP&A
  4. 04Strategic decision support
  5. 05Finance systems
AUTOMATIONhow the work runs
  1. 01Manual
  2. 02Structured
  3. 03Analytical
  4. 04Automated
  5. 05Intelligent
2.3 · IDENTITY
IDENTITY · RESOLVEDL01 → L04
Muhammad Ahmed Saeed
Muhammad Ahmed Saeed
FP&A Expert · Sunny'sFounder · Spot
MAAT · ACCA · BIDA® · FPAP®LAHORE, PAKISTAN · REMOTE · US / UAE TEAMS
RECORD■ HELD□ IN PROGRESS
EDUCATION
BSc (Hons) Applied AccountingOxford Brookes University
FIRST CLASS
DESIGNATION
MAAT (UK)Association of Accounting Technicians
MEMBER
QUALIFICATION
ACCAAssociation of Chartered Certified Accountants
EXPECTED 2027
CERTIFICATION
FPAC™Certified Corporate Financial Planning & Analysis Professional
CANDIDATE
OPERATES ACROSS
FINANCEDATAAUTOMATIONSYSTEMSINTELLIGENCE

Career model

LAYERS ACCUMULATE · STACK → NETWORK
L04 · Mar 2026 – PresentFP&A Expert · Sunny'sL03 · Dec 2023 – Mar 2026FP&A Analyst · VaporVML02 · Dec 2023 · dual roleAccounts & Finance Exec. · VaporVML01 · Dec 2023 – Feb 2024Intern · VaporVM
LAYERS 1/4 · DEPENDENCIES 0■ EVIDENCE PRESENT
LAYER 01Dec 2023 – Feb 2024Lahore, Pakistan

Accounts & Finance Intern · VaporVM

SYSTEMS
  • Led data collection and SOP creation for the Odoo accounting module rollout, making sure historical data imported correctly.
DATA
  • Analysed customer payment trends to improve collection strategies and reduce outstanding receivables.
LAYER 02Dec 2023 · dual roleDubai, UAE (Remote)

Accounts & Finance Executive · VaporVM

SCOPE
100%payroll accuracy
  • Payroll for roughly 120 employees (PKR 30–40M).
  • End-to-end invoicing; vendor and employee communications through delayed payment cycles.
DATA
  • Tracking dashboard for pending POs and aging analysis.
AUTOMATION
−85%petty-cash discrepancies
  • Introduced serialized petty-cash tracking.
OUTPUT
+15 daysvendor settlement terms
  • Weekly cash-inflow forecasts based on average receivable days.
  • Negotiated vendor payment terms, preserving working capital during tight periods.
LAYER 03Dec 2023 – Mar 2026Dubai, UAE (Remote)

FP&A Analyst · VaporVM

Dual role alongside accounts & finance.
SCOPE
  • Zero-based budgets for three departments (CS, MS, Telco), used to find allocation inefficiencies and target cost savings.
  • Owned the monthly management BDM MRR reporting structure.
COMPLEXITY
  • Performance comparisons over time, across service units and across BDMs.
SYSTEMS
  • Led development of new Odoo finance modules with external vendors and internal teams.
  • Built the company's Power BI analytics infrastructure from the ground up.
DATA
  • Pipeline from Odoo to Power BI; direct Odoo-to-Excel connection.
AUTOMATION
  • P&L statements, sales monitoring and target tracking automated.
  • Resource-utilisation, seating-capacity and overhead reports automated, cutting manual data entry.
OUTPUT
  • Revenue forecasting model in Python, using regression on the main business drivers.
  • Live year-over-year and budget-vs-actual variance views.
LAYER 04Mar 2026 – PresentVirginia, USA (Remote)

FP&A Expert · Sunny's

Reporting to the CEO and CFO.
SCOPE
  • Financial and profitability reporting for the executive team, built on a fleet-level P&L.
COMPLEXITY
  • Variance and variance-% against prior year, month, quarter, and 3- and 6-month trailing averages.
  • Multi-currency figures normalised against a daily FX calendar.
SYSTEMS
  • Star schema: separate revenue and cost facts joined through a shared reservation dimension; row-level security, reconciliations, documentation.
DATA
  • Power Query FX conversion from a live API; DAX month-to-date and year-over-year measures with complex filter-context logic.
AUTOMATION
  • Weekly and month-to-date management reporting via Power Automate, delivered as formatted email summaries.
OUTPUT
  • Margin root-cause analysis for the CEO and CFO, tracing a contribution-margin decline to a revenue-mix shift toward affiliate dispatch.
  • Waterfall bridges and a benchmarking view from segment down to individual vehicle.
Work with meDownload resume (PDF)
03 / SYSTEMSFINANCE OPERATING SYSTEM

Every tool here is tied to a result it produced.

My capabilities sit in six layers, from finance to AI. Pick one to see the tools I used for it and what the business got out of it.

CAPABILITY
L1 · FINANCE
L2 · ANALYTICS
L3 · DATA
L4 · AUTOMATION
L5 · SYSTEMS
L6 · AI
IMPLEMENTED WITH
ExcelPython · PandasPower BIDAXTableauSQLPower QueryFX rates APIPower AutomateOdoo ERPn8nPostgreSQLOpenAI GPT-4oQuickBooks OnlineXero
BUSINESS OUTPUT
Budget allocation inefficiencies surfacedRevenue projected from business driversMargin drivers identified and explainedExecutive profitability visibilityRecurring reports without manual prepControlled, reconciled accounting operationsDaily, auditable AR risk triageSpot · independent product →
SELECTED PATH · L1 FINANCEBudgeting & forecasting
IMPLEMENTED WITH → ENABLESExcel · Python · Pandas → Budget allocation inefficiencies surfaced · Revenue projected from business drivers
EVIDENCE · VAPORVMZero-based budgets for three departments (CS, MS, Telco); three-statement and rolling-forecast models.

04 / PROOFFINANCE CASE FILES

What the work produced, and how it was built.

Each case starts with a short summary. Open the blueprint to see the data, model and automation behind it. Where I have a measured result, it sits at the stage that produced it.

CASE FILE 01SUNNY'S · FP&A EXPERT
EVIDENCE · QUALITATIVE

Tracing a contribution-margin decline to its cause

BUSINESS PROBLEM
Contribution margin was declining and the executive team needed to know why.
INTERVENTION
Root-cause analysis on a fleet-level P&L, with waterfall bridges and segment-to-vehicle benchmarking.
OUTCOME
Decline traced to a revenue-mix shift toward affiliate dispatch; drivers and recommendations delivered to the CEO and CFO as an executive report.
CASE FILE 02VAPORVM · FP&A ANALYST
EVIDENCE · QUALITATIVE

Automating management reporting from the ERP

BUSINESS PROBLEM
Management needed performance comparisons over time, across service units and BDMs, and recurring reports relied on manual data entry.
INTERVENTION
Built the Power BI analytics infrastructure and a pipeline from Odoo, plus a direct Odoo-to-Excel connection for operational reports.
OUTCOME
Most management reports automated, with live year-over-year and budget-vs-actual views.
CASE FILE 03VAPORVM · ACCOUNTS & FINANCE
EVIDENCE · QUANTIFIED

Protecting working capital through tight payment cycles

BUSINESS PROBLEM
Tight cash periods, delayed payment cycles and petty-cash discrepancies.
INTERVENTION
Renegotiated vendor terms, introduced serialized petty-cash tracking, and built PO and aging tracking with weekly cash-inflow forecasts.
OUTCOME
Settlements extended by 15 days; petty-cash discrepancies cut by 85%; payroll for ~120 employees at 100% accuracy.
CASE FILE 04INDEPENDENT BUILD · n8n · ODOO · POSTGRESQL · GPT-4o
EVIDENCE · WORKFLOW BUILD

An AR workflow where AI interprets but never calculates

BUSINESS PROBLEM
Finance needs a daily view of which customers carry the most receivable risk and why, without handing the numbers to an AI model.
INTERVENTION
A scheduled n8n workflow pulls open receivables from Odoo, ages and scores every customer deterministically, and sends only high and critical accounts to GPT-4o for a structured interpretation.
OUTCOME
A daily, reconciled run: every source line ends as a valid line or a logged exception, and every AI output is validated before it reaches the database.

Credentials

SUPPORTING RECORD
EDUCATION & QUALIFICATION
OXFORD BROOKES UNIVERSITY · 2024BSc (Hons) Applied AccountingFirst Class Honours
MAAT (UK) · MEMBERAssociation of Accounting Technicians
ACCA · EXPECTED 2027Association of Chartered Certified Accountants
Nationwide 2nd · Performance Management · 87%Financial Accounting · 93%
FPAC · CANDIDATECertified Corporate Financial Planning & Analysis Professional
CERTIFICATIONS
  • BIDA® · Business Intelligence & Data AnalystFINANCE & DATA
  • FPAP® · FP&A ProfessionalFINANCE & DATA
  • Data Analysis in Excel SpecializationFINANCE & DATA
  • FP&A Excel Modelling SpecializationFINANCE & DATA
  • Business Intelligence Analyst SpecializationFINANCE & DATA
  • QuickBooks Online AccountantACCOUNTING
  • Xero AdvisorACCOUNTING
  • SQL AssociateTECHNICAL
05 / SPOTSTATE CHANGE · MODEL → PRODUCT

The same skills, applied to my own product.

My career work happens inside other companies' finance teams. Spot is the product I'm building myself. It's finance software that keeps the numbers exact where they have to be, and shows the uncertainty where they can't be.

DETERMINISTICf(x) = yMOVE ACROSS TO CHANGE x

Same inputs, same answer, every time. Ledgers, reconciliations and statements have to work this way.

PROBABILISTICP(y | x)MOVE ACROSS TO CHANGE x

How likely each outcome is, given what we know. Forecasts belong here, so Spot shows them as ranges with likelihoods, calculated on top of the exact core.

5.2 · SYSTEM UNDERNEATHDATA → CORE → PROBABILITY → AGENTS → DECISION
  1. 01Data inLedger, ERP and operational data, reconciled before anything else runs.
  2. 02Deterministic coref(x) = yRules, reconciliations and statements that always return the same answer.
  3. 03Probabilistic layerP(y | x)Forecasts and scenarios expressed as ranges with likelihoods.
  4. 04AgentsJames · Jane · Jason · Mark
  5. 05DecisionA recommendation someone can act on, with the uncertainty visible.
AGENT LAYER · IN DEVELOPMENT

James, Jane, Jason and Mark will be the names of Spot's AI agents. I'm still working out what each one will do.

  • James, Spot mascotAGENT 01 · JAMES
  • Jane, Spot mascotAGENT 02 · JANE
  • Jason, Spot mascotAGENT 03 · JASON
  • Mark, Spot mascotAGENT 04 · MARK
Meet SpotEarly access for finance teams, and updates for investors following the build.
→ OUTPUT

Work with me.

I take on FP&A, reporting and finance automation work. I'm based in Lahore and work remotely with teams in the US and UAE.

Contact me Download resume ↓
LinkedIn ↗