Ekki is a Product Manager who has spent 6 years in advanced technology products such as AI (Natural Language Processing and Computer Vision) and blockchain with diverse experience building products from 0 to 1.
While in Nodeflux, he successfully led AI product development and generated more than $4 million in the first year after launch. He also initiated a new innovative product called "Liveness Detection" to bring automation in E-KYC for fintech & banks, used by the top three banks and top fintech in Indonesia, impacting more than 20 million people's lives.
Currently leading the biggest NFT marketplace in the Near ecosystem, focused on product strategy and growth.
🏠 East Jakarta, Indonesia
📨 [email protected]
🖇️ linkedin.com/in/ekkirinaldi
⌨️ Telegram: @djinnamber
Notable Works
Achievements in AI
Work Experience
Artificial Intelligence Consultant @ Central Bank of Indonesia
Feb 2023 - Present (Contract)
- Help Bank Indonesia implement AI for fraud management.
Lead Product Manager @ Paras NFT Marketplace
2022 - Present
As IC
- Establishing Agile Product Management best practices with the development team.
- Closely collaborate with 100K MAU from the community and average over $4.000.000 monthly transactions.
- Rework marketplace design to be focused on buyer/demand. Bringing 121% more purchases on the first month of its release. Check on https://new.paras.id/
As Manager
- Designing company metrics to achieve alignment across teams.
- Design Product Team career track.
- Design company KPI
Group Product Manager @ Nodeflux
2021-2022
- Leading 3 AI Product Team: identifai.id, retailmatix.com, visionaire.ai.
- Create strategy for Identifai.id to create seamless onboarding solution for Digital Onboarding. Used by top 3 banks and top fintech in Indonesia.
- Supervise MLOps development.
- Monitor and evaluate metrics to measure the efficiency of the business plan.
Product Manager @ Nodeflux
2020-2021
- Launching a new AI Product - Visionaire. Gaining $ 4 million in revenue in the first year of launch.
- Initiate development of Liveness Detection. Shorten Digital Onboarding process from 24 Hours to just 15 minutes. Used by top 5 Banks and Fintech in Indonesia.
- Reduce Product Onboarding Time of product up to 95% and Image Size to 95,83% by redesigning product architecture from Python-based to C++ binary.
- Reduce Product Total Cost of Ownership (TCO) by up to 96,05% by changing platform, optimizing pipeline, and finding equilibrium between model performance and computation consumption.
- Increase up to 210% of the development performance by simplifying the product lineup, prioritizing backlogs, and aligning management vision with team capabilities.
Solution Architect @ Nodeflux
2018-2020
- Responsible for 26 projects from integration to end-to-end solution implementation.
- Working with a multi-national company to work on a $10 million project in Jakarta.
- Understand and synthesize customer needs from market research, customer feedback, analytics, competitor feature benchmark, and SWOT analysis.
- Lead implementation of AI and e-KYC solution in various banks.
AI Researcher @ Universitas Gadjah Mada
2018
- Research on sentiment analysis, inventing FVEC-CNN, deep learning method to classify the data with additional FVEC feature. Speaker in EECSI 2018 in Malang.
Product Manager @ GDP Labs
2017
- Research market and make market segmentation based on product.
- Help team to analyze each competitor and its bargain power.
- Develop idea to create new product to solve problem in the market
Education
Universitas Gadjah Mada - 2018/2021
Master of Computer Science
- Major in Artificial Intelligence / GPA 3.54
- Research in Computer Vision - Crowd Estimation Algorithm
Universitas Gadjah Mada - 2013/2017
Bachelor of Computer Science
- Major in Artificial Intelligence
- Research in Natural Language Processing
Publications
Speaker @ IEEE Conference - 18 October 2018
Mining opinions from Indonesian comments from YouTube videos are required to extract interesting patterns and valuable information from consumer feedback. This paper proposes new features FVEC and TF-IDF to represent the comments.
Participant @ IEEE Conference - 2 November 2017
This study classifies YouTube comments into seven classes. The experiment uses 13,638 Indonesian self-labeled comments of YouTube videos that review about smartphone products.
Skills
Product Development · Product Management · Computer Vision · NLP · Product Leadership · Docker · API · Python
Languages
Bahasa, English, French
Raising $ 4 Million Sales in the 1st Year: Reinventing AI Product to Fit the Market