# Bolaji BALOGOUN · Data Engineer · Data Architect (full content) > Data Engineer · Data Architect building production data platforms, operational data products, and governed automation. Single-file full version. Structured guide: https://www.bolablg.com/llms.txt · Biography: https://www.bolablg.com/bio.md · Interactive Q&A: https://chat.bolablg.com Equivalent job titles: Staff Data Engineer · Principal Data Engineer · Senior Data Engineer · Data Platform Engineer · Data Platform Architect · Data Architect · Cloud Data Architect · Data Engineering Manager · Head of Data · Director of Data Engineering · Data Director. Geographic scope: United States · five African operating countries · Gozem headquarters in Singapore. ## Biography # Bolaji BALOGOUN · Biography *Data Engineer · Data Architect.* Former Head of Data with a decade across data science and advanced analytics, including over seven years building production data pipelines and platforms. Bolaji combines Staff-level engineering depth with experience leading teams and data functions, from cloud migration and high-volume pipelines to operational data products, governance, and cost-disciplined architecture. ## Platform engineering and leadership Bolaji was Gozem’s first data hire. As Global Data Analyst, he built the analytical function from the ground up, migrated tens of terabytes to GCP and BigQuery through 50+ pipelines in one quarter without interruption, and delivered foundations for operations, finance, marketing, and fraud-pattern monitoring. He later led the data function as Head of Data, growing the team from 4 to 14+ and owning governance, FinOps, analytics engineering, and fintech data products. As Data Director through July 2026, he owned the data function end to end and co-led Gozem’s AI strategy. ## Data products at operating scale Bolaji built and deployed the [Gozem Data Hub](https://data.gozem.co), a governed platform serving 650+ employees, field agents, and partners across five African operating countries, with Singapore as headquarters. He personally coded 17 of its initial 19 microservices; two engineers coded one service each, with Bolaji deploying one and reviewing and fixing the other before deployment. He designed database schemas and persistence models, authored contribution documentation, and used live debugging to onboard engineers to service integration and CI/CD. The platform combined React, FastAPI, Nginx, MongoDB, Redis, NATS, BigQuery, Google Cloud Storage, Docker Compose, Artifact Registry, and shared Python utilities. The Data Hub powered application-serving pipelines, operational workflows, data exploration, and self-service analytics beyond dashboards. Bolaji implemented centralized authentication and routing, JWT and API-key access, scoped role-based permissions, service-to-service tokens, OTP and refresh-token hashing, and rate limiting. He also set up health checks, retries, stale-execution recovery, rollback controls, targeted test suites, security checks, and GitLab CI/CD gates. The work supported 250+ pipelines with quality checks, lineage, access controls, and service-level monitoring. ## Applied AI, automation, and reliability Bolaji co-led Gozem’s AI strategy and shipped 30+ production AI and automation tools across document intelligence, RAG support, fraud alerts, monitoring, and agentic workflows. He personally prototyped invoice extraction in Vertex AI Studio, testing prompts and context before building the production service with extraction validation, retries, and human review. The invoice workflow processed thousands of paper invoices and reduced the work from one to one and a half weeks to less than one day, saving approximately 40 hours per month for each Financing Manager. A real-time market monitoring program maintained an 88% rider match rate during a 48% demand surge and supported a 10.2% peak-event supply-rate lift. He also built application-ready data flows from MongoDB, BigQuery, Google Cloud Storage, and Sheets, plus a near-real-time MongoDB CDC pipeline that published change events through Pub/Sub, processed them in Dataflow, updated BigQuery, and supported real-time visualization in Looker Studio. These systems paired data products with operational services and decision workflows. ## FinOps and analytical serving As Head of Data, Bolaji redesigned Dataform dependencies, incremental models, shared marts, partitions, MERGE patterns, and assertions. He introduced job-metadata categorization and cost monitoring that reduced Google Cloud costs by 42.57% in one quarter while preserving data freshness and integrity. He also deployed Metabase and ClickHouse on AWS EC2, then used Airflow to replicate selected BigQuery tables and partitions into ClickHouse with use-case-specific history windows of three to 18 months for lower-cost analytical serving. ## Earlier work and advisory practice At Rintio, Bolaji built a production Hadoop and Apache Spark data stack with Elasticsearch to process hundreds of gigabytes of agricultural data from multiple sources. He developed crop-yield prediction models, primarily for maize with additional models for beans, achieving R² above 80% in most cases. The farmer-facing web-application MVP combined predictive features with Apache Superset visualizations and was still in testing when he left. He also delivered an executive data and AI training mission for Société Générale Maroc. As a Data Management Consultant with the International Trade Centre, he designed and delivered data-management and data-strategy training for 15 early-stage ECOWAS startups, covering market analysis, marketing campaigns, and team-performance tracking. He coached five founders individually, and later cohorts reused the training materials. He co-founded Takwimu LAB, whose low-resource African-language NLP work won the AI4D Africa Award in 2020 and reached the AfricaNLP workshop at EACL 2021. He is also Founding Vice-President of [iSheero](https://www.isheero.com/), where he designed and delivered a hands-on Google Cloud streaming module using Pub/Sub, Dataflow, BigQuery, and Looker Studio, and contributed to the review of Benin’s national AI and Big Data strategy. ## Credentials and working languages - Google Cloud Professional Data Engineer Certificate, Nov 2025 - Databricks Academy: Fundamentals accreditation and eight Knowledge Badges covering Lakeflow, data engineering DevOps, and Apache Spark, Sep–Oct 2026 ([credential wallet](https://credentials.databricks.com/profile/momboladjibalogoun708273)) - AWS Data Engineering Foundations Certificate, Aug 2026 - Apache Airflow Hands-on Certificate, Feb 2024 - MSc in Applied Statistics, Distinction, University of Abomey-Calavi, 2018 - BSc in Statistics and Econometrics, University of Abomey-Calavi, 2013 Bolaji is a native French speaker with full professional proficiency in English. He is based in Little Rock, Arkansas, United States, and is open to the job market. His Databricks, Delta Lake, Lakeflow, and Unity Catalog experience is through projects and labs. ## Contact - Email: [hello@bolablg.com](mailto:hello@bolablg.com) - LinkedIn: [linkedin.com/in/bolablg](https://linkedin.com/in/bolablg) - GitHub: [github.com/bolablg](https://github.com/bolablg) - Website: [www.bolablg.com](https://www.bolablg.com) ## Experience (full detail) ### Data Director · GOZEM · Apr 2025 - Jul 2026 Owned data platform engineering and internal data products (the Data Hub and 30+ automation tools) and co-led AI strategy across Gozem's footprint: five African operating countries with Singapore as headquarters. Mandate: 14+ data professionals · TG · BJ · CM · GA · CG · Gozem Data Hub - Gozem Data Hub live for 650+ people (employees, field agents, and partners) - 30+ production automation tools shipped across the group - 250+ pipelines under observability with automated quality checks #### Gozem AI Strategy knowledge base Result: 49 validated use cases catalogued; 6 in implementation, with near-full top-management adoption of daily AI assistance. Problem: Gozem needed a coherent AI strategy spanning operations, logistics, driver management, ecommerce, vehicle financing, customer experience, and fintech, backed by a structured view of the AI vendor and tooling landscape. Solution: Documented internal AI needs across 9 business units (Transport, Ecommerce, Financing, Finance, Data, Product, Marketing, Operations, Sales) with 50+ stakeholders contributing. Conducted a structured market scan covering 133 tools and vendors across 38 categories: AI vendors, foundation model providers, and automation platforms relevant to Gozem's stack. Catalogued 49 validated use cases. Built an adoption framework around the 6 moved into implementation: content creation and daily assistance (near-full adoption by top management), coding and code review (rolling out with the product team), classifier and others (in use by the data team). Each use case ships with its agentic workflow architecture. Tools: Data collection and analysis · Internal stakeholders interviews · Market data scraping · LLM-powered tools profiling · Strategic framework documentation #### Gozem Data Hub: Central data & AI workspace Result: Architected and built the platform, personally coding 17 of its initial 19 microservices; the governed workspace served 650+ employees, field agents, and partners across 5 countries. Problem: Business and operational teams alongside some external partners across Gozem's countries lacked a unified workspace for data and AI applications with fast iteration. Solution: Architected and built the Data Hub, personally coding 17 of its initial 19 microservices. Two engineers each coded one service; I deployed one and reviewed and fixed the other before deployment. Designed database schemas, persistence models, centralized authentication, routing, and scoped access controls. Beyond the initial microservices, documented the shared architecture and contribution process, then used live debugging to onboard engineers to service/frontend integration, CI/CD, and code review. The platform supported data-driven applications, operational workflows, automation, data exploration, and self-service analytics for people collaborating across Benin, Togo, Gabon, Cameroon, and Congo. Tools: React · FastAPI · Nginx · MongoDB · Redis · BigQuery · Google Cloud Storage · NATS · Docker Compose · Artifact Registry · Shared Python utilities #### Application-serving data pipelines Result: Delivered application-ready datasets for internal operational applications, services, and microservices, extending the platform beyond BI outputs. Problem: Operational tools required governed, application-ready data from systems with different storage patterns, not dashboard-only extracts. Solution: Designed Airflow-orchestrated pipelines sourcing MongoDB, BigQuery, Google Cloud Storage, and Google Sheets. Modeled the database schemas and persistence layers for data-intensive services, including data generated by application use and operational workflows. Tools: Apache Airflow · MongoDB · BigQuery · Google Cloud Storage · Google Sheets · Data modeling #### Platform delivery and reliability controls Result: Established repeatable service delivery with health monitoring, recoverable job execution, security gates, staged images, and rollback to prior images. Problem: A multi-service operational platform needed consistent release and runtime controls without overstating test coverage or disaster-recovery maturity. Solution: Implemented service health checks, dependency and restart controls, execution logs, configurable retries, and stale-execution recovery. Established GitLab CI/CD with linting, targeted pytest suites, security and container scans, multi-architecture image builds, Artifact Registry publication, staged promotion, and rollback to previously published images. Tools: GitLab CI/CD · Docker · Artifact Registry · Pytest · Trivy · Prometheus #### Governed KPI reporting & data-warehouse MCP Result: Business users gained a shared path to KPI reports built from documented metric definitions and canonical warehouse queries. Problem: Business users and AI-assisted workflows needed consistent KPI definitions, source tables, transformation logic, and query paths across the warehouse. Solution: Documented validated metrics with their source tables, transformation logic, and canonical queries; exposed the data warehouse through an MCP server and reusable BigQuery/Dataform views consumed by Looker Studio. This gave business users a shared reporting path grounded in documented sources. Tools: BigQuery · MCP · Python · Dataform · Looker Studio #### Automated invoice data extraction Result: ~40 hours saved per financing manager per month across thousands of invoices. Problem: Financing Operations team manually entered data from supplier and operational invoices in multiple formats, slowing reconciliation across Gozem's multi-country operations. Solution: Prototyped the invoice-extraction use case in Vertex AI Studio, testing prompts and context before production development. Architected and built an automated pipeline ingesting thousands of invoices per month (PDF and image formats) using LLM-powered OCR and document intelligence. Designed an agentic workflow supporting custom schema extraction, with validation to flag anomalies, retry handling, and human review, integrated into Google Sheets for reconciliation. Saved an average of 40 hours per financing manager per month. Tools: Vertex AI Studio (prototyping) · LLM OCR · Document Intelligence · Python · Cloud Storage · Google Sheets #### RAG-powered customer support chatbot Result: Resolvable support requests handled autonomously; complex cases escalated to Zendesk with conversation summaries preserving context. Problem: Customer support faced rising request volumes, with human agents spending too much time on resolvable cases and losing context when complex cases needed escalation. Solution: Oversaw development of a multi-agent Retrieval-Augmented Generation chatbot handling Gozem support requests autonomously, routing resolvable cases to the chatbot and escalating complex cases to Zendesk. Added automatic conversation summarization to preserve context for human handoff. Tools: RAG · Vector DB · Python · Zendesk #### Real-time market monitoring system Result: +10.2% supply rate, +7.4% request-to-completion, +5.4% coverage at the Vodun Days festival peak (~2,000 users). Problem: Transportation operations had no live visibility on supply-demand imbalances during high-demand events (public holidays, major urban events), leading to slow operational response. Solution: Architected a minute-granularity pipeline aggregating GPS, booking, and driver-availability signals. Built supply-demand imbalance detection with threshold-based alerts, plus an operational dashboard surfacing demand-density maps, supply-gap indicators, driver engagement scores, and event KPIs. Implemented automated Google Chat alerts and validated during the Vodun Days festival peak: 10.2% improvement in supply rate, 7.4% increase in request-to-completion rate, and 5.4% coverage gain across ~2,000 distinct users. Tools: MongoDB · BigQuery · Python · Apache Airflow · Looker Studio · Google Chat API - Architected and deployed **Gozem Data Hub**, a centralized platform serving over 650 people (internal employees, field agents, and external partners) with self-service analytics, AI and automation products, and data-driven applications for operations and business decisions. - Built line-of-business and operations apps (driver inspection, invoice extraction, external-acquisition agent performance monitoring, custom dashboard publisher, and others) used daily across Gozem's operations. - Built G-Force, a mobile-first web application giving external acquisition agents visibility into submissions, validation status, rejection reasons, and payments, with approximately twice-daily synchronization. - Co-shaped Gozem's AI strategy and shipped 30+ production AI and automation tools spanning fraud alerts, data extraction, and market monitoring; introduced AI-assisted engineering workflows to leadership and product engineering. - Built observability across 250+ pipelines (Dataform, Airflow, cron jobs) with automated data-quality checks, discrepancy alerts catching incomplete data and duplicates, performance tracking on exceptional events, and a data-exploration MCP. ### Head of Data · GOZEM · Oct 2022 - Mar 2025 Owned the data function for Gozem's Sub-Saharan Africa group: governance, team scaling from 4 to 14+, FinOps, and fintech-grade analytics. #### Dataform modernization and BigQuery FinOps Result: Reduced Google Cloud costs by 42.57% in one quarter while maintaining data freshness and integrity expectations. Problem: Scheduled BigQuery queries had grown into a costly transformation estate with limited dependency control, repeated computation, and weak visibility into which workloads drove spend. Solution: Led the migration to production Dataform modeling with explicit dependencies, incremental models, shared intermediate marts, partition-aware MERGE logic, and scoped assertions. Analyzed BigQuery job metadata, separated ingestion, transformation, reporting, and direct-analysis workloads, then targeted the models, schedules, and queries responsible for avoidable cost. Tools: BigQuery · Dataform · SQL · BigQuery job metadata · Incremental modeling #### Production orchestration on self-hosted Airflow Result: Replaced cron-based orchestration with a production scheduler supporting dependency-aware execution and operational monitoring. Problem: Cron jobs could trigger recurring work, but did not provide the dependency management, recovery controls, and shared operational view required as the pipeline estate expanded. Solution: Designed and deployed self-hosted Apache Airflow on AWS EC2 as the production orchestrator, migrated scheduled workloads from cron, and used DAG-level dependencies, retries, and monitoring to operate data workflows consistently. Tools: Apache Airflow · AWS EC2 · Python · Cron #### Lower-cost analytical serving with ClickHouse Result: Created a bounded-history serving path for recurring dashboards, reducing direct analytical pressure on BigQuery. Problem: Interactive Metabase dashboards repeatedly querying the central warehouse increased BigQuery demand even when each use case needed only a limited history window. Solution: Deployed open-source Metabase and ClickHouse on AWS EC2. Built Airflow replication workflows that moved selected BigQuery tables or partitions into ClickHouse with use-case-specific history windows ranging from three to eighteen months. Tools: ClickHouse · Metabase · Apache Airflow · BigQuery · AWS EC2 #### Data team capacity expansion Result: 12 hires onboarded end-to-end from 50+ evaluated candidates, with 30/60/90-day ramp milestones. Problem: The data team capacity needed to scale across African operations to support growing operational and BI needs, requiring 2 senior hires alongside a data engineer owning pipelines and platform. Solution: Authored the job scopes, technical requirements, and business cases justifying the senior hires to the leadership. Designed and ran technical interviews and assessments, evaluated 50+ candidates, negotiated offers end-to-end, and coordinated onboarding for 12 total hires with tooling access, team integration, and 30/60/90-day ramp milestones. Tools: Team capacity planning · Hiring pipeline management · Team productivity tools · Candidate evaluation · Onboarding #### Data science use-case knowledge base Result: 100+ use cases catalogued; top 15 adapted with Gozem-specific feasibility and outcome assessments. Problem: The data team needed a curated reference of proven use cases to prioritize roadmaps, drawing inspiration from how Gojek, Grab, and Uber operate at scale and adapting to Gozem's Sub-Saharan markets. Solution: Catalogued 100+ use cases across demand forecasting, driver scoring, fraud detection, surge pricing, and route optimization. Selected the top 15 and adapted each with Gozem-specific feasibility, data availability, and applicability assessments. Defined success metrics, technical challenges, infrastructure requirements, and estimated business value, structured as a living reference for the data and product teams. Tools: Research & development · Market analysis · Use-case design · Proof of concept development #### Ebanking analytics and reporting infrastructure Result: ~150 incidents surfaced per month; flagged a fraud worth more than 25M XOF; fintech warehouse powering regulatory reporting. Problem: Gozem Money (fintech vertical) needed real-time fraud monitoring, a fintech-grade data warehouse, and a reporting layer to support mobile payment operations and regulatory needs across African markets. Solution: Designed and supervised deployment of a micro-batch processing pipeline monitoring anomalous patterns, fraud signals, and threshold breaches. The system surfaces ~150 incidents per month and flagged a major fraud worth more than 25M XOF. Built the fintech data warehouse for Togo operations, powered by ETL/ELT pipelines and data models as the foundational layer for financial analytics, regulatory submissions, and stakeholder dashboards (transaction volumes, product adoption, revenue, cohorts). Established data quality pipelines and documented data models, reporting glossary, and metric definitions. Tools: Apache Airflow · Python · MongoDB · BigQuery · Dataflow · Cloud Storage #### Self-service no-code data exploration platform Result: 12+ active manager users; routine query load on the central data team halved. Problem: Non-technical users across operations, finance, product, and customer support depended on the central data team for routine reporting and ad-hoc queries, creating a bottleneck. Solution: Assessed data literacy and exploration needs across all Gozem markets. Coordinated configuration of a no-code query and exploration layer (Metabase) integrated with Gozem's central data warehouse. Adoption reached 12+ active manager users, halving routine query volume on the central data team. Tools: Metabase · BigQuery · Looker Studio #### Delivery agent incentive automation Result: Incentive-related ticket volume cut nearly in half; support disputes resolved within 24h. Problem: Delivery-agent positioning across high-demand zones was unmanaged, and incentive payments required manual finance intervention plus heavy customer-support load to resolve disputes. Solution: Modeled demand by zone, time-of-day, and event type to identify underserved areas. Designed a performance framework (acceptance rate, completion rate, zone presence, quality scores) and end-to-end payment automation from incentive calculation through disbursement instructions. Built a justification database with pre-computed data points enabling support to resolve cases within 24h. Cut incentive-related ticket volume nearly in half through pre-emptive payment automation. Tools: BigQuery · Python · Airflow · Google Sheets - Built and mentored the data science and analytics team from 4 to 14+ across Sub-Saharan Africa, establishing data catalog practices and a domain-oriented data governance framework with domain leads. - Led FinOps initiatives delivering a 42.57% reduction in Google Cloud costs in a single quarter through data model redesign, BigQuery warehouse schema optimization, orchestration optimization, cost monitoring, and ETL/ELT pipeline tuning, while maintaining SLA on data freshness and integrity. - Led the development of ML-based fraud screening and financial transaction monitoring pipelines, reducing field agent payment processing time by 7× and delivering a Group Financial Closing Database for enterprise-wide performance monitoring. - Co-developed a domain-oriented data governance framework with the team, delivering a KPI framework with definitions, data source documentation, and table descriptions across the group. - Co-chartered the marketing analytics stack (CAC, LTV, cohort, and multi-touch attribution embedded into weekly Growth reviews), owning the end-to-end supporting data pipelines from campaign platforms. - Integrated analytics and ML features into mobile apps and internal platforms with Product and Business teams, including dynamic surge pricing for transportation and an extension of the agent incentive automation use case. ### Global Data Analyst · GOZEM · Nov 2019 - Sep 2022 First data hire: built Gozem's analytics function from the ground up: GCP/BigQuery migration, monitoring automation, fraud-pattern detection, and marketing segmentation. #### Cloud migration to GCP & BigQuery warehouse Result: Migrated tens of terabytes in one quarter without interruption; the BigQuery warehouse and 50+ scheduled pipelines became the analytical foundation. Problem: The data infrastructure ran on legacy batch processes that couldn't keep up with growing operational and analytical needs across countries. Solution: Led the cloud migration of the entire data infrastructure to GCP, designing a BigQuery-based data warehouse with 50+ scheduled ETL pipelines and migrating tens of terabytes within a quarter without interruption. Tools: GCP · BigQuery · Python · Cron · Scheduled BigQuery queries #### Operational monitoring automation Result: Replaced a 2-day manual job with a 5–15 minute automated cycle for operational scenarios. Problem: Operations monitoring relied on a manual recurring job to spot business scenarios across markets, taking up to 2 days of analyst work per cycle and creating detection lag. Solution: Automated the operational monitoring workflow within the first 6 months of the role. Python jobs querying PostgreSQL surfaced supply gaps, driver behaviour anomalies, fraudulent trip patterns, and other scenarios into Google Sheets for the operations team, replacing a 2-day manual job with a 5-15 minute automated cycle. Tools: Python · PostgreSQL · Google Sheets #### Incentive payment automation Result: Reduced a multi-day manual payment process to minutes, with structured logs and dispute trails for finance and support. Problem: The agent and partner incentive payment process took multiple days, with manual reviews creating delays, errors, and a heavy load on the finance team. Solution: Built automated data pipelines for incentive payments combining structured operational data and near-real-time monitoring. Reduced a multi-day manual process to minutes, with structured logs and dispute trails for finance and support. Tools: BigQuery · Python · Airflow · Looker Studio #### Fraud patterns database & daily monitoring Result: Surfaced collection anomalies the same day they appeared and established groundwork for later ML-based fraud screening. Problem: Financial defaults from undetected fraud patterns were eroding margins, with no central knowledge base of known schemes or systematic daily collection monitoring. Solution: Implemented a fraud patterns database fed by automated daily collection monitoring, surfacing anomalies the same day they appeared and laying the groundwork for the later ML-based fraud screening pipeline. Tools: BigQuery · Python · Looker Studio #### Marketing segmentation & cohort analysis Result: Gave Marketing structured rider and driver profiles, cohorts, and campaign signals for targeted analysis. Problem: Marketing campaigns were broadly targeted because the team lacked a structured user profiling layer, leading to low engagement and inefficient spend. Solution: Built a profiling database covering riders and drivers, with RFM segmentation (recency, frequency, monetary) and cohort analysis used by Marketing for targeted campaigns. Translated cohort behavior into operational signals across retention, monetization, and re-engagement, supporting rider retention and growth. Tools: BigQuery · Python · Looker Studio #### Product and marketing data ingestion Result: Centralized product events, experiment data, and advertising performance in BigQuery for recurring product and campaign decisions. Problem: Product and Marketing teams lacked a shared analytical path across application behavior, A/B tests, campaign delivery, and audience profiles. Solution: Built ingestion workflows from Firebase and Google Analytics, Google Ads, and Facebook/Meta Ads into BigQuery. Modeled the data for product analytics, A/B-test analysis, campaign performance, user profiling, and product-development decisions. Tools: Firebase · Google Analytics · Google Ads · Facebook/Meta Ads · BigQuery #### Seasonal driver-retention analysis Result: Supported an approximately 9% improvement in driver retention through season-aware incentive decisions. Problem: Driver availability and retention varied with the regional agricultural season, but incentive planning did not account for the timing and strength of that effect. Solution: Analyzed driver activity against the regional agricultural calendar, identified the seasonal retention pattern, and translated the findings into incentive recommendations for Operations. Tools: BigQuery · Python · Cohort analysis · Google Sheets #### Business health monitoring & daily insights Result: Consolidated 20+ KPIs into a daily report used by country managers and top management for operational decisions. Problem: Country managers and top management lacked a single daily lens on business health across markets, with no consolidated KPI rhythm or recurring insight delivery. Solution: Built business health dashboards in Google Sheets, transitioning to Looker Studio and Tableau after the BigQuery migration. Consolidated 20+ KPIs into a daily report spanning all business areas. Presented insights daily to country managers and top management, surfacing trends and anomalies for operational decisions. Tools: Google Sheets · Looker Studio · Tableau · BigQuery - Led cloud migration to GCP, replacing legacy batch processing with a BigQuery-based data warehouse and scheduled ETL pipelines. - Built more than 50 Python workflows during the Global Data Analyst period for analytical and operational automation serving Operations, Finance, and Marketing, separate from the migration's 50+ ETL pipelines. - Built automated data pipelines for incentive payments, reducing a multi-day manual process to minutes through structured operational data and near real-time monitoring. - Implemented a fraud patterns database with automated daily collection monitoring, surfacing anomalies the same day they appeared and laying groundwork for later ML-based fraud screening. - Founded the analytics engineering function and grew a distributed team of 4, delivering automated dashboards, KPI reporting, and weekly market insights. - Refined Marketing segmentation and cohort analysis through a user profiling database, supporting targeted campaigns and retention analysis. ### Data Scientist · RINTIO · Jul 2018 - May 2019 Built data infrastructure and crop-yield forecasting for Duniya, Rintio's farmer-facing smart-farming MVP for Beninese farmers. #### Big data platform for precision agriculture Problem: Agricultural data arrived in varied formats and needed a unified storage and compute layer for analysis and product development. Solution: Designed and deployed a production data stack combining a Hadoop data lake, an Apache Spark cluster, and Elasticsearch. Processed hundreds of gigabytes of agricultural data from multiple sources and built ingestion and transformation pipelines to produce ML-ready features. Tools: Hadoop · Apache Spark · Elasticsearch · Python #### Crop yield forecast models Problem: The farmer-facing smart-farming MVP required crop-yield forecasts based on agricultural data. Solution: Developed regression-based crop-yield forecast models, primarily for maize with additional models for beans. Achieved R² above 80% in most cases. Tools: Python · scikit-learn · Apache Spark #### Farmer-facing smart-farming web-application MVP Problem: The product needed to present crop-yield predictions alongside agricultural data visualizations for farmers. Solution: Developed crop-yield models for a farmer-facing web-application MVP that combined predictive features with Apache Superset visualizations. The application was still in testing when I left. Tools: Apache Superset #### Data & AI executive training for Société Générale Maroc Problem: Société Générale Maroc's top management needed structured exposure to data and AI applications in banking to inform strategic priorities. Solution: Designed training content and developed lab use cases tailored to banking applications of data and AI, delivered to Société Générale Maroc's top management in a single training mission. Tools: Training content design · Lab use case development · Python - Designed and deployed a production data stack using Hadoop, Apache Spark and Elasticsearch, processing hundreds of gigabytes of agricultural data from multiple sources into ML-ready features. - Built regression-based crop-yield forecast models, primarily for maize with additional models for beans; achieved R² above 80% in most cases. - Developed crop-yield models for a farmer-facing smart-farming web-application MVP combining predictive features with Apache Superset visualizations; the application was still in testing when I left. - Delivered an executive data & AI training mission to Société Générale Maroc's top management, covering banking-specific use cases and hands-on labs. ### Statistician · INStaD · Sep 2017 - Feb 2018 #### R training programme for agency staff Problem: Agency departments needed practical exposure to advanced analytics tools but lacked accessible, hands-on training material tailored to their day-to-day workflows. Solution: Designed and delivered hands-on R analytics training to ~30 agents from 2 departments over a 1-week seminar (2 sessions). Left agency staff with reusable training materials. Tools: R · RStudio · Markdown #### Survey funding data collection at public ministries Problem: Benin's national research financing knowledge base was incomplete, particularly on survey-related funding flows across public institutions, limiting visibility for policy and planning. Solution: Contributed to INStaD's national survey funding data collection campaign by covering the statistics departments of 4 ministries. Collected and reported research financing records to enrich the agency's national knowledge base. Tools: R · Excel · Field data collection - Designed and delivered hands-on R analytics training to ~30 agents across 2 departments over a 1-week seminar. - Contributed to a national survey funding data collection campaign at INStaD, covering 4 ministries' statistics departments. ### Statistician · Le PANTAGRUEL · Aug 2013 - Oct 2013 #### Customer satisfaction survey & scoring system Problem: Sales had no structured feedback mechanism: client perceptions were anecdotal, NPS untracked, and there was no way to translate satisfaction signals into operational decisions. Solution: Designed and conducted a customer satisfaction survey with the Sales team, delivering a structured client feedback database. Built and deployed a satisfaction scoring system, then trained Sales to use it for ongoing performance reviews. Findings shaped management policies that lifted NPS by approximately 9 points. Tools: Excel · Survey design · Statistical scoring - Designed and conducted a customer satisfaction survey with the Sales team, delivering a structured client feedback database. - Findings translated into management policies that lifted NPS by approximately 9 points. - Built and deployed a customer satisfaction scoring system, then trained the Sales team to use it in performance reviews. ## Other experiences (advisory · community) ### Data Management Consultant · International Trade Centre - ITC (UN Agency) · 2023 As a Data Management Consultant, designed and delivered data-management and data-strategy training for 15 early-stage ECOWAS startups under the UN ITC Netherlands Trust Fund V program. - Created reusable training materials covering market analysis, marketing campaigns and team-performance tracking to support startup competitiveness and internal operational efficiency. Later cohorts reused the materials. - Delivered the workshops to 15 startups from Benin and other ECOWAS countries, and coached five founders individually on their data-management challenges. ### Founding Vice-President of iSheero · iSheero · Dec 2021 - Present iSHEERO is an initiative promoting data science and AI in Benin and across Francophone Africa, built on three pillars: - Skills development: equipping young Beninese talent with AI skills. - Local ecosystem: building a data science ecosystem to position Benin as a hub for international AI projects. - AI in business: helping companies adopt AI in their operating workflows. Contributions as Founding Vice-President: - Designed and delivered a hands-on Google Cloud real-time streaming module using Pub/Sub, Dataflow, BigQuery, and Looker Studio at the Africa TechUp Tour bootcamp. - Engaged with DataCamp scholarship recipients in data engineering as part of an iSHEERO partnership program. - Contributed to the review of Benin's national big data strategy roadmap, with the Ministry of Digital Affairs. - Co-trained directors of information systems (DSIs) across Ministry of Finance departments in data science. To learn more about our activities, events and team, visit our website. ### Chair of Takwimu LAB · Takwimu LAB · 2019 - 2022 Co-founded Takwimu LAB, a community for AI and Data Science learners in Francophone Africa, building hands-on programs from online learning paths. Led research on low-resource African languages, collecting, labeling, and publishing parallel text datasets for Neural Machine Translation (French-Fongbé, French-Ewe). Built a prototype English-Fongbé Neural ML Translation app through the AI4D-Africa Network with UNESCO support. Recognition includes the Zindi Africa Award for the AI4D-African Dataset Challenge (2020). Co-organized a machine translation challenge on Zindi Africa. Published findings via the AI4D African Language Program, presented at the AfricaNLP Workshop during EACL 2021. - Parallel text dataset for Neural Machine Translation (French-Fongbé, French-Ewe) published November 2020: https://zenodo.org/record/4266935 - AI4D African Language Program, April 2021. Accepted and presented at the AfricaNLP Workshop, EACL2021 Conference. arXiv:2104.02516 [cs.CL] ### Big Data Bootcamp · RINTIO Big Data Bootcamp · Mar - Jun 2018 Intensive 4-month program covering Data Engineering, Machine Learning, and Big Data infrastructure (Spark, Hive, Python, SQL), with hands-on projects on data visualization and large-scale processing. Key project: **Duniya Data**, a collaborative agricultural insights app applying data science to farming practices, built with peers under mentor guidance. ## Education ### MSc in Applied Statistics (Distinction) · University of Abomey-Calavi · 2015 - 2018 Focus areas: statistical modeling, survival analysis, statistical inference on large datasets, high-dimensional data processing. ### BSc in Statistics and Econometrics · University of Abomey-Calavi · 2010 - 2013 ## Credentials - Google Cloud Professional Data Engineer Certificate · Google Cloud · Nov 2025 - Stream Processing and Analysis with Apache Spark™ · Databricks Academy · Oct 2026 - Monitoring and Optimizing Apache Spark Workloads on Databricks · Databricks Academy · Oct 2026 - Developing Applications with Apache Spark™ · Databricks Academy · Oct 2026 - Introduction to Apache Spark 4 · Databricks Academy · Oct 2026 - DevOps Essentials for Data Engineering · Databricks Academy · Sep 2026 - Build Data Pipelines with Lakeflow Spark Declarative Pipelines · Databricks Academy · Sep 2026 - Deploy Workloads with Lakeflow Jobs · Databricks Academy · Sep 2026 - Data Ingestion with Lakeflow Connect · Databricks Academy · Sep 2026 - Academy Accreditation - Databricks Fundamentals · Databricks Academy · Sep 2026 - AWS Data Engineering Foundations Certificate · AWS Training & Certification · Aug 2026 - Apache Airflow Hands-on Certificate · Udemy · Feb 2024 - McKinsey.org Forward Program · McKinsey & Company · Jul 2025 ## Publications - Participatory Research for Low-resourced Machine Translation · AfricaNLP Workshop, EACL 2021 · Apr 2021 · https://arxiv.org/abs/2104.02516 - Parallel Text Dataset for Neural Machine Translation (French-Fongbé, French-Ewe) · Zenodo · Nov 2020 · https://zenodo.org/record/4266935 ## Recognition - CAF2025 speaker: Super Apps & Data Protection · On Gozem's super-app approach to data protection · 2025 - WakeUp Champ podcast guest · On the path from statistics to data science leadership · 2023 - Zindi Africa Award · AI4D African Dataset Challenge (Takwimu LAB) · 2020 ## Proof points - ~5 TB · Batch upserts/day (average) - 45M · Streaming events/day (average) - ~300 · Dataform models & Airflow DAGs - −42.57% · GCP cost in one quarter - 650+ · Data Hub employees, field agents & partners ## Tech stack - Data platforms & governance: BigQuery · ClickHouse · BigLake · Databricks · Dataplex · Unity Catalog - Databases & storage: PostgreSQL · MongoDB · Google Cloud Storage · AWS S3 · Delta Lake · Apache Iceberg - Pipelines & streaming: Airflow · Dataform · Dataflow · Apache NiFi · Lakeflow · Spark · Pub/Sub - Languages & frameworks: Python · SQL · R · Bash · JavaScript · Scala · React · FastAPI - Cloud & delivery: Google Cloud · Compute Engine · AWS · AWS EC2 · Cloud Run · Cloud Functions (team delivery oversight) · Docker · Artifact Registry · Git - Analytics & visualization: Data Studio · Looker · Superset · Metabase · Plotly · Grafana - AI & automation: Vertex AI (team delivery oversight) · Gemini · Codex · RAG · LangGraph (familiarity) · MCP ## Contact - Email: hello@bolablg.com - LinkedIn: https://linkedin.com/in/bolablg - GitHub: https://github.com/bolablg - Blog: https://blog.bolablg.com - Based in: Little Rock, AR, US (America/Chicago) - Availability: Open to market.