1. Challenge Name: AI-driven financial companion

2. Context

In the digital era, customers expect banks to be not just a place to store money, but a companion that truly understands them. Vietcombank currently possesses the VCB Digibank digital banking ecosystem with a large amount of customer data and a solid technology platform. We are moving towards an “Empathetic Banking” model – where services go beyond payment transactions to focus on financial health care, being present as a capable assistant at every milestone in customers’ lives.

3. Pain Points & Problems to Solve

Despite possessing large amounts of data, customer interactions on VCB Digibank still face limitations:

• Lack of personal context: The system has not fully leveraged behavioral data to identify which life stage customers are in, leading to product suggestions that remain too generic.
• Lack of planning tools: Customers do not have intuitive tools to set up and track long-term financial goals (buying a car, getting married, studying abroad…) directly in the application.
• Passive interaction: Spending alerts or financial advice do not yet function as a proactive “virtual assistant,” failing to provide timely reminders when customers “overspend” or neglect their savings plans.

4. Desired Solution

Vietcombank is seeking AI/Big Data solutions to build a VCB LifeMap module integrated into VCB Digibank, including core capabilities:

• Analysis & Understanding
  • Use AI to analyze transaction history and spending behavior to automatically build LifeMap.
  • Classify cash flows and accurately identify customers’ current life stages.
• Virtual Assistant & Advisory
  • AI Chatbot acting as an advisor: Alert when spending exceeds thresholds, remind about savings progress.
  • Suggest appropriate financial products (savings, investments, insurance, loans) for each life stage.
• Goal Planning
  • Allow customers to set specific financial goals with real-time timelines.

5. Technology & Implementation Requirements

• Technology: AI/Machine Learning applications with the ability to train on large datasets, ensuring high accuracy and continuous updates.
• Integration: The solution must integrate seamlessly (via API/SDK) into the VCB Digibank ecosystem without disrupting the current experience.
• Security & Compliance: Strictly comply with personal data protection regulations (Decree 13) and banking information security; internally processed data must not generate credit risk.
• MVP Roadmap: Ability to build a pilot version (MVP) for internal customers (Vietcombank employees) within 6 months before scaling up.

6. Evaluation Criteria

Solution proposals will be evaluated based on a combination of creativity and quantitative performance indicators:

A. Business Impact:

• Conversion Rate: How much is the solution expected to increase the cross-sell conversion rate of financial products compared to traditional methods?
• Portfolio Share: Ability to increase the average number of products used per customer.

B. Customer Engagement Metrics:

• Interaction Frequency: Does the solution increase the frequency of customers accessing Digibank (beyond transfer transactions) for the purpose of tracking financial plans?
• Retention Rate: Ability to retain users who continuously use LifeMap features after 3-6 months.

C. Technical Metrics:

• Model Accuracy: Accuracy rate in identifying “Life Stage” and classifying spending behavior (Target: >95%).
• Response Speed: Ensure the ability to process and provide suggestions with low latency to avoid affecting user experience.

D. Feasibility:

• Time-to-market: Complete MVP for internal testing within a maximum of 12 months.
• Scalability: System architecture ready for mass deployment to millions of Digibank users after the MVP phase.

We sincerely thank you for your cooperation! This detailed information will serve as a solid foundation for creating breakthrough AI solutions.

PART 1: CONTEXT & CHALLENGES

1. Current Situation

Many enterprises in Vietnam are facing workforce shortages and shifts, especially in rapidly changing industries (technology, finance, services).

HR data is scattered across multiple systems (HRIS, payroll, internal surveys, performance evaluations), but is rarely analyzed in depth to support decision-making.

Workforce planning mainly relies on experience and manual spreadsheets, lacking AI tools to support analysis and forecasting.

2. Issues to Address

Enterprises lack the ability to forecast future workforce demand and compare it with current capacity, leading to difficulties in recruitment, training, and talent retention.

Current analytical tools only provide descriptive reports, not supporting proactive decision-making.

There is no integrated system to help HR and leadership have a “panoramic view” of the workforce: from attrition risks, capability gaps, to HR cost allocation.

3. Urgency

In the context of rapidly changing economy and labor market, the ability to forecast and respond quickly is a critical competitive advantage.

Building a Workforce Intelligence Platform will help enterprises not only optimize resources, but also enhance employee experience and business performance.

PART 2: OBJECTIVES & SOLUTION REQUIREMENTS

4. Core Objectives of the Solution

Stating the most important objectives that the solution needs to achieve

• Develop an AI platform to support enterprises in proactive strategic workforce planning.
• Identify gaps between current workforce and needs for the next 2–5 years, then provide action recommendations.
• Predict attrition risks, skill shortages, and suggest appropriate recruitment/training strategies.
• Integrate existing HR data to create a visual “Talent Heat Map” for managers.

5. Key Results to Achieve

Please share the desired output results. E.g.: Business effectiveness, operational efficiency, scalability and integration,…

• Enterprises can monitor and forecast workforce supply-demand by department/skill/job position.
• Reduce annual workforce planning time by 20–30% compared to manual methods.
• System provides early warnings about HR risks (e.g., departments at risk of critical workforce shortage in the next 6–12 months).
• Solution can integrate with existing HRIS/ERP systems (SAP, Oracle, Workday, or domestic HRIS).
• Visual reports, easy to use for leadership and HR departments.

PART 3: DIRECTION & SUPPORTING RESOURCES

6. Technology Direction

Does the enterprise have any specific direction?

• Encourage the use of AI/ML to forecast and analyze workforce trends.
• Can combine RAG (Retrieval Augmented Generation) to leverage labor policy and industry data.
• Solution should be developed as a SaaS/B2B platform, scalable for multiple countries and languages.
• Use MCP (Model Context Protocol) model to connect enterprise systems.

7. Supporting Resources from the Enterprise (if possible)

What resources can the enterprise provide to potential startups that the enterprise wishes to collaborate with?

• Talentnet can provide anonymized HR data (aggregated, anonymized) for startups to test models.
• Provide expert mentorship in HR (HR consulting, compensation & benefits, labor law).
• Support pilot testing opportunities with clients in Talentnet’s network.
• Connect with corporations and large enterprises with actual Workforce Intelligence needs.

We sincerely thank you for your cooperation! This detailed information will be a solid foundation for creating breakthrough AI solutions.

AI CONTACT CENTER SOLUTION

PART 1: CONTEXT & CHALLENGES

1. Current Situation

• Contact Center inbound service is facing a large volume of repetitive queries accounting for 40–60% of total interactions.
• The Voice of Customer and NPS collection process is still manual, time-consuming, does not accurately reflect actual customer needs, and depends on personal judgment.
• There is no comprehensive customer data analysis tool to predict behavior or proactively suggest engagement.
• The call center team handles a large volume of calls manually, affecting performance, consulting quality, and limited upsell capability due to lack of necessary information about customer purchasing behavior, mainly relying on experience and recommendations from the fieldforce.

2. Issues to Address

• Lack of intelligent automation (AI automation) in inbound/outbound to optimize productivity.
• Fragmented data, difficult to integrate between CRM systems, hotline, Zalo OA, and e-commerce channels, manual order placement and order recording steps affecting work efficiency.
• No AI-powered quality control (QC) monitoring tools leading to subjective evaluation, not strict enough and inconsistent.
• Lack of real-time consumer behavior analysis capability to suggest suitable products, increase upsell/cross-sell.
• Sales capability differences between individuals lead to results heavily dependent on consulting ability, monitoring, and easily influenced by personal emotions in sales activities.

3. Urgency

• Competition in the FMCG industry is increasingly fierce, customers demand fast response, personalization and omnichannel.
• Competitors in banking, telecommunications, and e-commerce have successfully deployed AI Contact Center with operating costs reduced by 30–40%, productivity increased by 20–30%, and NPS increased by 5–10%.
• AI integration becomes a mandatory condition to maintain competitive advantage, reduce personnel costs, and flexibly scale services in the context of limited resources.

PART 2: OBJECTIVES & SOLUTION REQUIREMENTS

4. Core Objectives of the Solution

Stating the most important objectives that the solution needs to achieve

• Increase operational efficiency through AI automatically handling 30–50% of call volume and repetitive interactions.
• Improve service quality & NPS with faster, more personalized responses, 24/7.
• Optimize costs & headcount, reduce operating costs by 20–25% through schedule optimization and automated monitoring.
• Indirectly increase revenue through intelligent AI product recommendations and identification of upselling/cross-selling opportunities.
• Standardize & improve agent training quality through AI QC & speech analytics.

5. Key Results to Achieve

Please share the desired output results. E.g.: Business effectiveness, operational efficiency, scalability and integration,…

Objective	Desired Result	Target Timeline
Reduce repetitive request volume	20–30%	Q3/2026
Increase agent productivity	+5–10%	Q4/2026
Optimize personnel budget	~10%	Q4/2026
Improve average NPS	+5 points	Q4/2026
Increase indirect order volume	+5.5K HLs	Q4/2026

PART 3: DIRECTION & SUPPORTING RESOURCES

6. Technology Direction

Does the enterprise have any specific direction?

AI Layer integrating Contact Center Platform (Genesys & Microsoft Dynamics, Zalo OA, External AI platform/module):

Create an intermediate AI layer (AI Hub) connecting customer data, transaction history, and interaction channels to automatically respond, suggest actions, and intelligently route.

Main AI Applications (Anticipated Use Cases):

1. AI Inbound Automation: Automatically handle common requests, integrate intelligent IVR & Zalo OA chatbots.
2. Conversational AI for Telesales: Support agents in real-time decision making, suggest SKUs, and personalize sales scripts.
3. AI Quality Monitoring: Analyze 100% of calls, automatically detect process deviations & suggest training.
4. Predictive Engagement: AI predicts churn risk or repurchase potential, activate proactive care campaigns.

Data Infrastructure & Integration:

• CRM Cloud (Microsoft Dynamics)
• Speech Analytics (Genesys, Google Speech-to-Text API, Open AI, Gemini)
• Unified Data Lake serving reporting & machine learning models

7. Supporting Resources from the Enterprise (if possible)

What resources can the enterprise provide to potential startups that the enterprise wishes to collaborate with?

From internal enterprise:

• Contact Center Transformation project team (RTC, D&T, Marketing, Finance)
• Customer history data, call records, tickets, call logs, chat logs

From AI partners & APAC AI Hub:

• Technology consulting & deployment roadmap
• Support for pilot models and quantitative KPI measurement
• Financial/budget support in the initial phase to ensure scalability

We sincerely thank you for your cooperation! This detailed information will be a solid foundation for creating breakthrough AI solutions

AI-DRIVEN NET ZERO 2030: SUPPLY CHAIN TRANSPARENCY SOLUTION

PART 1: CONTEXT & CHALLENGES

1. Current Situation

Vietnam’s textile and garment industry is entering a critical transformation phase as major export markets (EU, USA, Japan) simultaneously require supply chain data transparency and disclosure of product sustainability indicators. However, most small and medium-sized enterprises (SMEs) still lack the digital tools and capabilities to implement multi-tier traceability, measure environmental impact, or standardize data according to international standards (DPP, PEFCR, CSRD…).

2. Issues to Address

How can enterprises like Faslink and their partner ecosystem:

• Automate the collection, standardization, and transparency of supply chain data;
• Measure and optimize sustainability indicators for each product;
• Apply technology in a simple – suitable – scalable manner for SMEs in the value chain.

3. Urgency

From 2026, regulations such as Digital Product Passport (EU) and EPR Textile will be implemented, requiring transparent data at the product level. Without early preparation, Vietnamese enterprises risk losing export opportunities and failing to achieve the national Net Zero 2030 target. Therefore, applying AI in traceability and sustainability measurement is urgent to standardize and accelerate digital transformation in the industry.

PART 2: OBJECTIVES & SOLUTION REQUIREMENTS

4. Core Objectives of the Solution

Stating the most important objectives that the solution needs to achieve

• Build an AI tool to support enterprises in automatically collecting, analyzing, and standardizing supply chain data (from yarn – weaving – dyeing – garment manufacturing – logistics).
• Measure sustainability indicators (CO₂, water, energy, recycling, lifecycle) and suggest optimization towards Net Zero 2030 goals.
• Design a solution that is simple, cost-effective, and compatible with SMEs, easily integrated with internal systems and international standards (PEF, DPP, GHG Protocol).

5. Key Results to Achieve

Please share the desired output results. E.g.: Business effectiveness, operational efficiency, scalability and integration,…

• Reduce supply chain data collection and standardization time by 70%.
• Build automatic transparent reports (traceability + sustainability dashboard) for products.
• Support Faslink and SME partners to pilot implementation in 2026, towards a comprehensive sustainable-transparent operating model by 2030.
• Enhance international reputation and competitiveness through data transparency technology.

PART 3: DIRECTION & SUPPORTING RESOURCES

6. Technology Direction

Does the enterprise have any specific direction?

Faslink prioritizes practical AI solutions with the capability to:

• Recognize and classify data from multiple formats (files, invoices, QR codes, certificates);
• Automatically analyze environmental indicators and provide improvement recommendations;
• Create Digital Traceability Profiles according to DPP standards, integrable with ERP systems or internal dashboards.

7. Supporting Resources from the Enterprise (if possible)

What resources can the enterprise provide to potential startups that the enterprise wishes to collaborate with?

• Real data from Faslink’s yarn fabric, dyeing, and garment manufacturing supply chains and partner network.
• Testing infrastructure (sandbox) at Faslink GLAB – material R&D and innovation center.
• R&D team and technical advisors to accompany startups during solution development and testing.
• Market connections with over 200 fashion – garment – brand partners domestically and internationally to expand application after testing.

We sincerely thank you for your cooperation! This detailed information will be a solid foundation for creating breakthrough AI solutions.