AI PRODUCT ENGINEER & DATA SCIENTIST

Hi, I'm Farhan.

From complex
problems to
intelligent systems.

I connect business understanding, AI and engineering to build software that solves real problems.

Explore my workDownload CV

Jakarta, Indonesia Ideas → working products

Farhan Zahirsyah, wearing glasses and a black shirt
A story of building useful things
BUSINESS THINKING.
ENGINEERED INTO REALITY.

02THE WAY I WORK

The right solution starts
with the right problem.

From the first question to a working system, I connect the business intent with the details that make it work.

01 / DISCOVER

Understand the real problem.

I start with stakeholders, users and the constraints around their work. A clear problem gives the system a useful purpose.

Business discovery→Clear requirements→Operational software

03END-TO-END PRODUCT ENGINEERING

The product.
The platform.
The whole system.

I architect and build across the frontend, backend, data and integration layers—connecting the parts into one useful product.

GoPython / FastAPIVuePostgreSQLRedis

Select a module to explore the engineering.

An integrated product systemIllustrative architecture based on CV responsibilities: CRM, RAG agent and Meta integrations connect through Go/FastAPI services, PostgreSQL, Redis, and Vue frontend. Twenty CRMLeads · records · follow-upsAI agentRAG · tools · human handoverMeta integrationsWhatsApp · marketing APIsBackend & API servicesGo · Python / FastAPI · async processingPostgreSQLStructured application dataRedis queueBackground jobsWorkspace · illustrative UIVue frontend / customer portal

Illustrative architecture based on my work at PT Miliki Rumah.

04AI + CRM + AUTOMATION

A conversation
becomes action.

AI is most useful when it can work with the systems around it. I build agents that retrieve context, use tools, update records and know when to involve a person.

RAGTool callingTwenty CRMWhatsApp

A synthetic customer journey. No messages are sent.

CUSTOMER JOURNEYSIMULATION
Hi, can you help me explore a mortgage?
AI assistant

I can help explain the options and collect the details a specialist needs.

Relevant knowledge retrieved RAG
DEMO LEADDemo visitor
Stage
New inquiry
Next step
Understand the need

01 / A new inquiry enters the system.

05DATA SCIENCE & FINANCIAL INTELLIGENCE

Find the signal.
Make a better decision.

Document analysis, credit scoring and reporting connect raw information to decisions people can review.

Financial intelligence with reviewable signalsSynthetic bank statement with a layout signal, illustrative credit-scoring inputs, and KPR/RTO/human-review routing. No real model outputs or customer data. Bank statement analysisSynthetic statement · document signalsDEMO BANK / STATEMENTSample data · January 2026DateDescriptionAmountJan 02Salary transfer+8,500,000Jan 05Household payment−1,250,000Jan 09Cash withdrawal−2,000,000Jan 14Transfer−700,000Jan 18Utility payment−450,000Layout / metadata signalA review signal requires investigation.It does not establish transaction fraud.Credit scoring & routingIllustrative inputs and outcomesIncomeCommitmentsFinancial historyMODELExampleEligibility routingSynthetic examples · not live decisionsKPRMortgage routeRTOAlternative routeReviewHuman assessment
SYNTHETIC EXAMPLE

Review signals.
Human judgment.

PDF metadata and layout anomalies can identify documents that need investigation. A formatting signal alone does not establish transaction fraud.

Ready to inspect the sample document.

06BUSINESS ANALYSIS & SOLUTION DESIGN

Make the thinking clear.
Then make it work.

I translate business objectives into explicit requirements, workflows and interfaces.

→→
From requirements to productEditable requirements document, intermediate wireframe and product UI. The three groups can be morphed or repositioned independently. RequirementsPRD / FRDBusiness objectiveUser journeyData modelSuccess conditionsWireframe & workflowStructureWorkspace · illustrative UIImplemented interfaceBusiness objectives → explicit workflows → working product

Business objective, user flow, data model and success conditions—made explicit before implementation.

Trading-platform experience includes requirements and stakeholder coordination across KYC, deposits, withdrawals and back-office workflows.

07THE ENGINEERING LAB

The right tools.
Working together.

01 / DATA & ANALYTICS

Cells become
business clarity.

SQL, spreadsheets and automated reporting bring structure to operational data.

SQLSheetsApps ScriptLooker Studio
fx=SUM(C2:C6)
Editable synthetic calculation example
RowItemValueStatus
2Sample A✓
3Sample B✓
4Sample C✓
5Sample D✓
6Sample E✓
Total1,075Σ

Select a cell to inspect it, or recalculate the sample.

02 / BACKEND & INFRASTRUCTURE

A request.
A coordinated system.

Services, databases and background workers connect business logic to reliable application workflows.

PythonGoFastAPIPostgreSQLRedisDocker
GET/api/demo/customer
RouteRequest validation
ServiceBusiness logic
DatabaseSample record
WorkerBackground job
Ready
{ "demo": true, "status": "ready" }

Local simulation; no network request is sent.

03 / FRONTEND & PRODUCT

Interfaces that
complete the system.

Clear workflows, responsive screens and connected application state turn the underlying engineering into something people can use.

VueTypeScriptREST APIsProduct workflows
Product workspace
OverviewConnected
Records128Workflows06Updates24

Illustrative interface · synthetic values

08PROJECT POSSIBILITIES

Ideas worth building.

Four product worlds. Choose one, step inside, and try the workflows for yourself.

Pick a product. Explore it hands-on.

Continue the story

Sample data · Browser-local prototypes · Preview illustrations are synthetic

09CAREER & QUALIFICATIONS

Different chapters.
One building mindset.

Data science, financial technology, business analysis and end-to-end product engineering.

Master of ManagementWidyatama University · GPA 3.96/4.00Bachelor of Industrial EngineeringPasundan University · GPA 3.47/4.00Hacktiv8 Data Science Program2022
Jun 2022Project-Based Data ScientistID/X PartnersLoan risk scoring, feature engineering and predictive modeling.
  • XGBoost and Random Forest models.
  • Data preparation, feature engineering and evaluation.
Explore this work
Jul 2022 — Jun 2025Business Data AnalystPT Samuel Kripto IndonesiaTrading-product requirements, reporting automation and customer analytics.
  • PRD/FRD across KYC, deposits, withdrawals and back-office workflows.
  • SQL, Google Apps Script, Sheets and Looker Studio reporting.
  • Financial and regulatory reporting pipelines.
Explore this work
2024LSP Pasar Modal · CertificationRegistered Technical AnalystEarned during the Samuel Kripto tenure.
  • Financial-market technical analysis certification.
Jun 2025 — PresentData Scientist (AI Product & Solutions Engineering)PT Miliki RumahEnd-to-end AI software, CRM, integrations and financial intelligence.
  • Integrated Go, Python/FastAPI, Vue, PostgreSQL and Redis systems.
  • RAG chatbot, tool calling, WhatsApp and Twenty CRM.
  • Credit scoring, KPR/RTO routing and document analysis.
Explore this work

10LET’S MAKE IT USEFUL

Let's build
the next useful thing.

Have a problem worth solving, an idea to explore,
or a team building something meaningful?

Start a conversationDownload CV
Emailfarhanzahirsyah1@gmail.comLinkedInLet’s connectGitHubExplore the code
me@farhanzahirsyah.my.id·farhanzahirsyah.my.id

ENGINEERING NOTES

Based on my responsibilities at PT Miliki Rumah.