I turn data and complex problems into better products.

The Problem

Monthly and weekly reporting at Bongo required significant manual effort. Pulling data from multiple platforms, formatting dashboards, and distributing reports consumed 20+ hours per week across the analytics team.

Context

As the platform scaled across OTT, MCN, and Studio ecosystems, the volume of stakeholder-facing reports grew faster than the team. Manual processes became a bottleneck for timely decision-making.

My Role

Led the automation initiative end-to-end: identified bottlenecks, designed automation workflows, wrote Python and R scripts for data pipelines, and built self-updating dashboards in Looker Studio.

Approach

1. Audited existing reporting workflows to identify repetitive manual tasks
2. Designed automated data pipelines using Python and R
3. Built self-refreshing dashboards in Looker Studio connected to live data sources
4. Created standardized report templates to ensure consistency
5. Rolled out automation incrementally, training team members on the new system

Tools & Technology

Outcome

Reduced manual reporting effort by 40–50%, saving the team approximately 20 hours per week. Dashboards became available 24/7 with real-time data, enabling faster stakeholder decisions.

What I Learned

• Automation is most effective when you standardize the process before automating it.
• Stakeholder buy-in comes from showing time savings in the first week, not promising them.
• Self-serve dashboards reduce ad-hoc reporting requests significantly.

The Problem

Content across Bongo's platforms (OTT, YouTube, Facebook) lacked a unified identification system. The same piece of content had different IDs on different platforms, making cross-platform analytics unreliable.

Context

Without standardized Content IDs, teams could not accurately track content performance across platforms, leading to fragmented insights and conflicting reports to stakeholders.

My Role

Proposed the Content ID standardization project, designed the matching logic, coordinated with engineering and content teams, and drove 100% stakeholder adoption.

Approach

1. Mapped content catalogues across all platforms to identify naming inconsistencies
2. Designed a standardized Content ID schema that could work across OTT, YouTube, and Facebook
3. Built matching algorithms to reconcile existing content with the new ID system
4. Worked with engineering to integrate the standard into platform databases
5. Presented the system to stakeholders and iterated based on feedback

Tools & Technology

Outcome

Achieved 85% match accuracy across platforms and 100% stakeholder adoption. Cross-platform analytics became reliable for the first time, enabling unified content performance reporting.

What I Learned

• Data standardization is a product problem, not just a technical one. You need active buy-in from every team that touches the data.
• Starting with the highest-traffic content first builds momentum and demonstrates value quickly.

The Problem

Mojo features at Bongo needed a structured product development process to translate business requirements into clear technical specifications, ensuring smooth handoff between data insights, product requirements, and engineering execution.

Context

As Bongo expanded its product portfolio, the gap between what stakeholders wanted and what engineering built was growing. Feature delivery was slower than needed, and alignment across teams was inconsistent.

My Role

Owned the data-to-product handoff for Mojo features. Wrote PRDs, defined user stories, facilitated grooming sessions, and worked directly with engineering to ensure accurate implementation.

Approach

1. Gathered requirements from stakeholders and translated business needs into product specifications
2. Wrote detailed PRDs and user stories with clear acceptance criteria
3. Facilitated grooming and sprint planning sessions with engineering and QA
4. Created data-backed prioritization frameworks for feature decisions
5. Monitored feature rollout metrics to validate product decisions

Tools & Technology

Outcome

Improved feature-to-user alignment and delivery time by 25%. Team velocity increased by 35% through structured grooming and planning processes.

What I Learned

• Clear acceptance criteria in user stories eliminates most back-and-forth during development.
• Data-backed prioritization makes stakeholder alignment faster because you argue with evidence, not opinions.
• Regular grooming sessions are the single highest-ROI process investment for product teams.

The Problem

A promotional website was needed before a scheduled press/event deadline for the "Salahuddin Ayyubi" project. Traditional development timelines would not meet the deadline.

Context

The project required rapid execution, compressing a complete website design and development cycle into hours instead of days. This demanded a combination of product thinking, clear design direction, and effective use of modern AI tools.

My Role

Led the entire project from Figma design to deployment, using AI-assisted development with Claude Code and prompt engineering to achieve a full website build in approximately 3 hours.

Approach

1. Created the visual design in Figma, establishing layout, typography, and brand identity
2. Used Claude Code and structured prompt engineering to accelerate development
3. Iteratively refined the output through targeted prompts and manual adjustments
4. Tested across devices and browsers before deployment
5. Deployed before the required press/event deadline

Tools & Technology

Outcome

Delivered a complete, responsive promotional website in approximately 3 hours, well before the event deadline, demonstrating the viability of AI-assisted product execution for rapid prototyping and deployment.

What I Learned

• AI-assisted development is most powerful when the human brings clear product vision and design direction.
• Prompt engineering is a product skill: structuring effective prompts is essentially writing clear micro-specifications.

The Problem

Understanding user behavior and creating meaningful audience segments was critical for content strategy and marketing decisions, but the team lacked a structured segmentation framework.

Context

With millions of users across digital media platforms, a one-size-fits-all approach to content and marketing was leaving value on the table. Different user segments had different content preferences and engagement patterns.

My Role

Designed and executed customer segmentation analysis using statistical methods and data visualization, translating complex datasets into actionable business segments.

Approach

1. Collected and cleaned user behavior data from multiple platform sources
2. Applied statistical analysis and segmentation methodologies
3. Created cohort analyses across OTT, Facebook, and YouTube platforms
4. Built visualization dashboards to make segments interpretable for non-technical stakeholders
5. Presented findings with specific content and marketing recommendations for each segment

Tools & Technology

Outcome

Delivered actionable user segments that informed content strategy and marketing campaigns. Cohort analysis provided ongoing visibility into user retention and engagement patterns across platforms.

What I Learned

• The best segmentation is the one that non-technical stakeholders can actually use to make decisions.
• Visualizations that tell a story drive action; tables of numbers get ignored.