AWS Financial Services Singapore

Crypto.com

Real-time Analytics

However, developers encountered challenges with the limitations of open-source models and the high cost of self-hosting large language models (LLMs). They also faced accuracy issues with outputs generated by open-source models, especially when dealing with multilingual news sites.

Read the original story Explore all use cases

Crypto.com is a crypto exchange and comprehensive trading platform serving 100 million users in 90 countries.

Crypto.com uses Anthropic Claude 3 large language models on Amazon Bedrock for sentiment analysis and application development, and Amazon SageMaker to fine-tune its custom models.

With the top models on AWS, the company can deliver accurate sentiment analysis in less than 1 second, and efficiently train and adjust new models in a rapidly evolving market.

Within one month, had implemented Anthropic Claude 3 Haiku models on Amazon Bedrock for sentiment analysis, collecting and analyzing crypto news in more than 25 languages.

Reported impact

Business outcome

By implementing a multi-agent consensus-seeking solution for sentiment analysis on AWS, can efficiently deliver accurate, comprehensive, and localized crypto market insights to its global user base. Fok explains, "We can share more instant, updated news with users on the sentiment of coins—whether they’re bullish or bearish, for example. Key reported metrics include second to return results from large language models 1; month to integrate Claude 3 models; languages Delivers localized multilingual content 25.

Solution

Crypto.com uses Amazon Bedrock with Amazon SageMaker Studio to run an efficient architecture that delivers nuanced, domain-specific crypto market insights to 100 million global users. Crypto.com is a crypto exchange and comprehensive trading platform serving 100 million users in 90 countries. To improve the service quality of Crypto.com, the firm implemented generative artificial intelligence (AI)-powered sentiment analysis services on AWS to generate market insights. Crypto.com uses Anthropic Claude 3 large language models on Amazon Bedrock for sentiment analysis and application development, and Amazon SageMaker to fine-tune its custom models. With the top models on AWS, the company can deliver accurate sentiment analysis in less than 1 second, and efficiently train and adjust new models in a rapidly evolving market. Source data enters the Crypto.com workflow through its application services, processed with Amazon SageMaker Studio, Amazon SageMaker, Amazon Bedrock.

Data flow

Source data enters the Crypto.com workflow through its application services, processed with Amazon SageMaker Studio, Amazon SageMaker, Amazon Bedrock.

Solution architecture

8 components · 1 layer

Compute
- Amazon SageMaker Studio Gives engineers a managed environment for ML experimentation and development.
- Amazon SageMaker Supports model training, fine-tuning, evaluation, or managed inference for the customer workload.
- Amazon Bedrock Hosts the foundation models and inference APIs that power the generative AI features.
- Anthropic Claude 3 Serves as the primary large language model for generation, analysis, or reasoning tasks.
- Amazon EC2 Provides configurable compute capacity for application, training, or inference workloads.
- Anthropic Claude 3 Haiku Serves as the low-latency large language model for generation or analysis tasks.
- Mistral AI Supplies open-source or commercial foundation models used in the solution.
- Meta Llama Supplies open-source foundation models that the customer fine-tunes or serves.

Architecture clues

Raymond Lam, senior engineer at , says, “Amazon SageMaker provides the tools and APIs to easily adapt our models with a user-friendly interface.
Sunny Fok SVP, Head of AI Innovation Technology at Crypto.com Crypto.com is a crypto exchange and comprehensive trading platform serving 100 million users in 90 countries.
With the top models on AWS, the company can deliver accurate sentiment analysis in less than 1 second, and efficiently train and adjust new models in a rapidly evolving market.
Within one month, had implemented Anthropic Claude 3 Haiku models on Amazon Bedrock for sentiment analysis, collecting and analyzing crypto news in more than 25 languages.
“We’re becoming more efficient with ready-to-use models we can access via API, which gives us more flexibility in development,” Lam says.

Evidence from the source

Crypto.com is a crypto exchange and comprehensive trading platform serving 100 million users in 90 countries.

Crypto.com uses Anthropic Claude 3 large language models on Amazon Bedrock for sentiment analysis and application development, and Amazon SageMaker to fine-tune its custom models.

However, developers encountered challenges with the limitations of open-source models and the high cost of self-hosting large language models (LLMs). They also faced accuracy issues with outputs generated by open-source models, especially w

These LLMs are highly scalable and process vast amounts of data in real-time, facilitating comprehensive market research.

With the top models on AWS, the company can deliver accurate sentiment analysis in less than 1 second, and efficiently train and adjust new models in a rapidly evolving market.

Within one month, had implemented Anthropic Claude 3 Haiku models on Amazon Bedrock for sentiment analysis, collecting and analyzing crypto news in more than 25 languages.

Transparency Evidence & methodology 95/100 reference value · 87% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value95/100

Problem clarity20/20
Measures whether the customer's operational problem is specific enough to compare.
Implementation depth20/25
Rewards named components, data flow, integrations, and architecture detail.
Outcome specificity20/20
Measures whether the reported benefit is concrete rather than a generic claim.
Quantified impact25/25
4 metrics describe measurable impact.
Deployment scale10/10
Rewards evidence of a live or scaled deployment and meaningful integration scope.

Evidence confidence87%

Source authority20/25
The main evidence is an official vendor-published customer story.
Claim traceability21/25
10 of 12 key fields link to a named source.
Direct evidence20/20
6 direct source quotes retained.
Date provenance8/10
The publication date comes from publisher-controlled metadata.
Source transparency8/10
1 source reference document publisher, URL, authority, retrieval, and supported fields.
Implementation specificity10/10
Rewards source-backed components, data flow, quantified impact, and a specific outcome.

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

AWS
Crypto.com customer story

First-party
https://aws.amazon.com/solutions/case-studies/case-study-crypto/

Source type
Vendor-published customer story

Published
Jul 5, 2024
Supports
- Customer
- Business problem
- Solution
- Products
- Solution components
- Data flow
- Integrations
- technical area
- use case category
- Business outcome
- outcome category
- Metrics
- Architecture
- evidence quotes
- Publication date

Date evidence

Publisher catalog date Jul 5, 2024 · AWS · First-party
Open date source

Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Similar use cases

See all like this

AWS Feb 14, 2025 MarketReader Financial Services United States MarketReader is an artificial intelligence (AI) analytics platform providing the financial sector with data-driven explanations of real-time asset movement. During development, MarketReader experienced delays in data… 100 percent Real-time Analytics Amazon Managed Streaming for Apache Kafka · Amazon S3 · Apache Kafka Connect

AWS Apr 15, 2024 Trustly Financial Services Global As Trustly started growing rapidly, the fintech company wanted to modernize its data analytics capabilities and become more data driven. Trustly can also update dashboards that benefit from near real-time data every 15 minutes rather than three… Real-time Analytics Amazon Q · Amazon Q in QuickSight · Amazon QuickSight

In-house Oct 9, 2026 BBVA Financial Services Spain BBVA's trading desk needs to respond to institutional client requests for bond, interest-rate swap, and foreign-exchange quotes within milliseconds while balancing competitiveness, profitability, portfolio risk, and… within milliseconds Real-time AnalyticsRisk Management Atalaya · Bloomberg · Viriato