// D4T4 SC13NC3 C0NSULT4NT · Dubai, UAE · Hyderabad, India
I turn data into strategic advantage. 10+ years building AI systems that drive measurable transformation across organisations managing $100M+ in revenue. 11 years · $50M+ value delivered · 6 global organisations.
10+yrs_experience
$50M+value_generated
120+pipelines_live
6orgs_transformed
cat./profile.md
SYSTEM PROFILE
Independent data science consultant with 10+ years turning messy, siloed data into production ML systems and strategic AI advantages. Currently freelancing across MENA and South Asia.
Core Process
Spent a decade learning one thing: the best model is the one that gets used. That means earning trust before deploying code, building systems that are explainable to a board and reliable in production simultaneously.
From McDonald's data corruption to MAG Group's $4.2M intelligence platform — every engagement starts with the same question: what decision are we actually trying to change?
AI transformation — GenAI strategy and implementation
Advisory — data platform architecture, team building
Freelance engagements — data science, ML, analytics consulting
tail-f ./career.log
EXECUTION LOG
10 years. 6 organizations. Every role left the data infrastructure better than it found it.
ls-la ./demos/
LIVE DEMOS
Not case studies. Actual running models. Change the inputs — the output updates in real time.
PROC_01 // M4RK3T1NG_4TTR1BUT10N
ROAS Uplift Modeller
PySpark uplift model served via FastAPI. Drag the sliders to shift channel spend and watch predicted Return on Ad Spend recalculate live.
PySparkFastAPIGA4DockerSalesforce
LIVE SIMULATOR
Paid Search Budget$42K
Social Media Budget$28K
Email / CRM Weight0.6
PREDICTED ROAS—
VS BASELINE—
PROC_02 // G3N3R4T1V3_AI
NL → SQL Agent
Llama-3 + LangChain agent over Redshift. Select a business question — the agent writes the SQL, executes it, and streams the result.
Llama-3LangChainRedshiftPython
QUERY INTERFACE
PROC_03 // PR3D1CT1V3_M0D3LL1NG
CLV Predictor
BG/NBD + Gamma-Gamma model. Input customer signals and receive a 3-year Customer Lifetime Value prediction with segment classification.
scikit-learnBG/NBDPythonSalesforce
PREDICT CLV
3-YEAR CLV—
SEGMENT—
PROC_04 // R3T3NT10N_M0D3LL1NG
Churn Risk Scorer
Gradient-boosted classifier trained on RFM + behavioural signals. Adjust customer attributes and get a real-time churn probability with recommended intervention.
XGBoostPythonSnowflakeFastAPI
CHURN CLASSIFIER
Days Since Last Purchase45d
Support Tickets (90d)2
NPS Score7
CHURN PROBABILITY—
ACTION—
PROC_05 // PR1C1NG_1NT3LL1G3NC3
Dynamic Price Optimiser
Elasticity model + competitor signal fusion. Set your current price and market context — the model outputs the revenue-maximising price point.
scikit-learnPySparkBigQueryPython
PRICE OPTIMISER
Current Price ($)$120
Competitor Price ($)$135
Demand Index0.7
OPTIMAL PRICE—
REV UPLIFT—
PROC_06 // D3M4ND_F0R3C4ST1NG
Sales Demand Forecaster
Prophet + XGBoost ensemble trained on 3 years of FMCG retail data. Adjust category, seasonality strength and trend to get a 90-day forward demand prediction.
ProphetXGBoostPySparkBigQueryPython
FORECAST ENGINE
Baseline Units / Week840
Seasonality Strength0.6
Trend Index+1.2%
90-DAY FORECAST—
VS BASELINE—
grep-r "impact" ./results/
MEASURED OUTPUT
Every number is documented. Every claim is auditable.
sshsiddie.me// establishing secure connection...
Ready to build something that actually works
I typically respond within a few hours. Tell me the problem you're trying to solve — not the solution you think you need.
UAE: +971 58580 6565IND: +91 98850 65637STATUS:● AVAILABLE FOR ENGAGEMENTS
// TYPICAL ENGAGEMENT
Week 1–2 · Discovery: understand your data, team, and decision bottlenecks Week 3–6 · Build: production-grade ML pipeline or GenAI solution Week 7+ · Handover: documented, tested, your team can own it