This submit was co-authored with Krišjānis Kočāns, Kaspars Magaznieks, Sergei Kiriasov from Solar Finance Group
For those who course of id paperwork at scale—mortgage functions, account openings, compliance checks—you’ve possible hit the identical wall: conventional optical character recognition (OCR) will get you partway there, however extraction errors nonetheless push a big share of functions into guide overview queues. Add fraud detection to the combo, and the guide workload compounds.
Solar Finance, a Latvian fintech based in 2017, operates as a technology-first on-line lending market throughout 9 international locations. The corporate processes a brand new mortgage request each 0.63 seconds and delivers greater than 4 million evaluations month-to-month. In considered one of their highest-volume industries, with 80,000 month-to-month functions for microloans, roughly 60% of functions required guide operator overview. Solar Finance partnered with the AWS Generative AI Innovation Middle to rebuild the pipeline. Inside 35 enterprise days of handover, the answer was reside in manufacturing. The next timeline reveals the complete venture journey from kickoff to manufacturing launch.
Solar Finance venture timeline from kickoff to manufacturing
The venture moved by 4 milestones over 107 enterprise days. The AWS Generative AI Innovation Middle engagement ran 32 days from kickoff (August 26, 2025) to ultimate presentation (October 9, 2025), adopted by 26 days for technical handover (November 14, 2025). Solar Finance then took 35 enterprise days to maneuver the answer into manufacturing, together with a 14-day manufacturing freeze over the vacation interval (December 18 – January 7), and went reside on January 22, 2026.
On this submit, we present how Solar Finance used Amazon Bedrock, Amazon Textract, and Amazon Rekognition to construct an AI-powered id verification (IDV) pipeline. The answer improved extraction accuracy from 79.7% to 90.8%, lower per-document prices by 91%, and diminished processing time from as much as 20 hours to below 5 seconds. You’ll learn the way combining specialised OCR with giant language mannequin (LLM) structuring outperformed utilizing both software alone. You’ll additionally learn to architect a serverless fraud detection system utilizing vector similarity search.
The Identification Verification Problem
Solar Finance had constructed its first IDV automation in 2019 utilizing Amazon Rekognition and Amazon Textract. As the corporate expanded into creating areas, the system’s limitations grew to become arduous to disregard.
This area offered distinctive challenges with language and doc complexity. Processing paperwork in each English and a neighborhood language proved troublesome for conventional OCR techniques. The native language textual content stays underrepresented in conventional OCR coaching datasets, inflicting frequent extraction errors. Solar Finance additionally wanted to deal with 7 totally different ID sorts, every with totally different layouts and codecs.
The guide workload was primarily pushed by OCR errors. Of the 60% of functions requiring guide overview, roughly 80% of circumstances stemmed from mismatches between extracted data and customer-entered information. Critically, 60% of those mismatches had been OCR errors, not buyer errors. The remaining 20% of guide interventions associated to fraud detection flags.
Fraud detection added one other layer of complexity. About 10% of every day requests had been precise fraudulent functions. Fraudsters used comparable photographs with distinctive patterns to bypass primary controls whereas submitting a number of mortgage functions. Figuring out these patterns required time-intensive guide overview throughout quite a few photographs.
Value and velocity constraints blocked enlargement. The per-document price and roughly 3 full-time equivalents (FTEs) devoted to guide verification on this area alone meant the unit economics blocked enlargement into industries with lower-value microloans. Processing instances ranged from below 10 minutes for automated circumstances to twenty hours for guide critiques outdoors enterprise hours.
Answer overview
The AWS Generative AI Innovation Middle ran a 6-week proof-of-concept (September–October 2025) centered on one high-volume {industry}. The staff constructed two AI-powered options: an ID extraction system and a fraud detection system. Each had been deployed as a totally serverless structure on AWS.The answer makes use of the next key providers:
- Amazon Bedrock – For AI structuring and visible evaluation utilizing Anthropic’s Claude Sonnet 4, and vector era utilizing Amazon Titan Multimodal Embeddings.
- Amazon Textract – For main OCR textual content extraction from id paperwork.
- Amazon Rekognition – For fallback OCR, face detection, and face masking.
- Amazon S3 Vectors – For serverless vector similarity search in opposition to identified fraud patterns.
- AWS Step Capabilities – For orchestrating parallel fraud detection workflows.
- AWS Lambda – For serverless compute throughout each pipelines.
The next diagram illustrates the answer structure.
Solar Finance API structure displaying ID extraction and fraud detection routes
The structure exposes two API routes by Amazon API Gateway, with mortgage software information saved in Amazon Easy Storage Service (Amazon S3):
- `/extract-id` route (ID extraction). An AWS Lambda operate receives the ID picture and sends it to Amazon Textract for main OCR. If Amazon Textract returns low-confidence outcomes, the system falls again to Amazon Rekognition for OCR. The extracted textual content is then handed to Amazon Bedrock (Claude Sonnet 4), which buildings it into standardized JSON fields.
- `/detect-fraud` route (fraud detection). An AWS Lambda operate triggers an AWS Step Capabilities workflow that runs two checks in parallel:
- Background similarity — Amazon Rekognition masks the face from the selfie picture, then Amazon Bedrock Titan Multimodal Embeddings generates a vector illustration of the background. This vector is queried in opposition to Amazon S3 Vectors to search out matches with identified fraud patterns.
- Visible sample detection — Amazon Bedrock (Claude Sonnet 4) analyzes the picture for display picture artifacts and digital manipulation.
Each outcomes feed right into a Lambda-based threat evaluation operate that produces a mixed fraud rating as JSON.
- Fraud ingestion pipeline (proper facet). Confirmed fraud photographs are ingested from Amazon S3 by a Lambda operate. The pictures are processed by Amazon Rekognition for face masking, vectorized by Amazon Bedrock Titan Embeddings, and saved in Amazon S3 Vectors. This grows the reference database over time.
Conditions
To implement the same answer, you want the next:
Answer walkthrough
This part walks by the 2 core pipelines: ID extraction and fraud detection.
ID extraction pipeline
The ID extraction system didn’t arrive at its ultimate design on day one. The staff iterated by three distinct approaches over 4 weeks, and every failure pointed towards the following enchancment. The next diagram reveals how the pipeline advanced from a single Claude Sonnet 4 through Amazon Bedrock strategy at 61.8% accuracy to the ultimate multi-tier design at 90.8%.
ID extraction: evolution of approaches displaying three iterations from 61.8% to 90.8% accuracy
Strategy 1: Claude Sonnet 4 alone (61.8% accuracy). The staff’s first try despatched ID photographs on to Anthropic’s Claude Sonnet 4 through Amazon Bedrock and requested it to extract fields as JSON. The outcomes had been disappointing: 61.8% general accuracy, with ID quantity extraction at solely 43%. The core subject was the mannequin’s built-in security protocols for dealing with personally identifiable data (PII). Claude is educated to restrict processing of delicate PII discovered on id paperwork like driver’s licenses, passports, and nationwide IDs. When offered with actual ID photographs, the mannequin triggered these privateness safeguards and refused to extract data from some information, which straight impacted efficiency. Moreover, even when extraction succeeded, sure fields (like ID numbers) confirmed poor accuracy as a result of the mannequin prioritized security over exact character recognition on delicate paperwork.
The takeaway: whereas Claude excels at normal doc evaluation and OCR duties, its built-in privateness protections make it unsuitable for direct extraction from id paperwork containing PII.
Strategy 2: Amazon Textract + Claude structuring (85% accuracy). The breakthrough got here when the staff separated OCR from structuring. Amazon Textract dealt with uncooked textual content extraction from ID photographs. Claude Sonnet 4 then structured the output into 7 standardized fields: doc kind, date of delivery, identify, surname, center identify, ID quantity, and expiry date. This single change produced an 11.6% accuracy leap.
This strategy labored as a result of Amazon Textract, as a specialised OCR service, doesn’t have the identical PII refusal mechanisms as Claude, so it reliably extracted textual content from each ID picture with out triggering security protocols. As soon as the textual content was extracted, Claude may deal with what it does finest: clever structuring. Claude excelled at dealing with native language textual content with diacritical marks, inferring lacking data from context, and making use of document-specific extraction guidelines. These are duties that conventional OCR alone couldn’t deal with. By working with already-extracted textual content somewhat than uncooked ID photographs, Claude prevented its security constraints.
The takeaway: separating considerations allowed every software to function inside its design parameters: Amazon Textract for dependable OCR and Claude for clever structuring.
Strategy 3: Multi-tier OCR + validation (90.8% accuracy). The ultimate iteration added Amazon Rekognition as a fallback for photographs the place Amazon Textract struggled (usually low-quality scans, uncommon doc angles, or broken IDs) plus validation guidelines for ID quantity formatting, date standardization, and doc kind normalization.
The multi-tier structure works as follows. Amazon Textract handles main OCR. Amazon Rekognition offers backup extraction when Amazon Textract confidence is low. Claude buildings the mixed output, and validation guidelines catch formatting errors that slip by. ID numbers get padded to the right size primarily based on doc kind, and dates are standardized to YYYY-MM-DD format. These validation guidelines proved essential. They caught edge circumstances the place OCR extracted right characters however in inconsistent codecs.
The next chart reveals the weekly accuracy development throughout 585 take a look at photographs. The staff didn’t beat the baseline till Week 4, once they added Amazon Textract. Every iteration revealed new failure modes that knowledgeable the following architectural enchancment.
ID extraction: the journey to 90.8% accuracy displaying weekly progress
The takeaway: combining specialised OCR instruments (Amazon Textract + Amazon Rekognition) with LLM structuring (Claude) and validation guidelines beats utilizing a single software alone for doc extraction.
Fraud detection pipeline
The fraud detection system makes use of AWS Step Capabilities to run two detection strategies in parallel, then combines their scores right into a ultimate threat evaluation.
Visible sample detection. Claude Sonnet 4 through Amazon Bedrock analyzes submitted selfie photographs for indicators of fraud: display pictures (seen bezels, scan traces, moiré patterns), display glare and reflections, and digital manipulation artifacts. Photos scoring 85% confidence or larger are flagged. The system ignores regular traits like blur, compression artifacts, and customary cropping to cut back false positives. Display screen picture detection works properly, with 95%+ confidence on identified patterns.
Background similarity evaluation. This element catches fraud rings, that are teams of fraudsters submitting selfies from the identical location. The pipeline works in three steps. First, Amazon Rekognition masks faces to deal with the background. Then, Amazon Titan Multimodal Embeddings generates a 1024-dimensional vector of the background. Lastly, Amazon S3 Vectors searches for matches in opposition to identified fraud patterns.
The staff examined each text-based and visible embeddings for similarity search. Textual content embeddings (having Claude describe the background, then evaluating descriptions) achieved 91% accuracy however solely 27.8% precision and 21.7% recall. Visible embeddings carried out much better: 96% accuracy, 80% precision, and 52% recall.
Background similarity: visible options strategy displaying the pipeline and textual content vs visible embedding comparability
Threat evaluation. The scoring algorithm weighs visible sample detection (50%) and background similarity (50%) equally. Scores of 75+ point out high-confidence fraud, 38–74 point out medium confidence, and under 38 is classed as reliable. The parallel execution structure processes photographs in 3–5 seconds, down from 6–8 seconds when run sequentially.
Serverless structure
All the answer runs on AWS Lambda, AWS Step Capabilities, and Amazon API Gateway. This design lets the staff modify particular person Lambda capabilities, take a look at modifications instantly, and deploy updates with out downtime. This was essential throughout a 6-week engagement the place the strategy modified weekly.
Authentication makes use of Amazon Cognito with AWS SigV4 request signing. AWS WAF protects in opposition to widespread internet safety points. Knowledge is encrypted at relaxation with AWS Key Administration Service (AWS KMS) and in transit through TLS 1.2+. The infrastructure is outlined in Terraform and handed safety audits with 25 findings analyzed: 14 false positives, 9 justified exceptions, and a pair of deferred for manufacturing.
Outcomes
The proof-of-concept delivered measurable enhancements throughout accuracy, velocity, fraud detection, and price.
ID extraction efficiency
The system was evaluated in opposition to 585 ID photographs:
Metric
Baseline
New answer
Enchancment
Title
84.93%
87.72%
+2.79%
Date of delivery
81.25%
90.80%
+9.55%
Doc kind
78.43%
96.40%
+17.97%
ID quantity
74.32%
89.40%
+15.08%
Total accuracy
79.73%
90.80%
+11.07%
ID quantity extraction, beforehand the weakest area at 74.32%, improved by over 15 proportion factors. Doc kind classification reached 96.4%. Common processing time: 4.42 seconds per doc.
Fraud detection efficiency
The mixed end-to-end fraud detection pipeline (visible sample detection plus background similarity) achieved 81% accuracy with 59% recall and 83% specificity.
Fraud detection outcomes: 81% accuracy, 59% recall, 83% specificity
The 59% recall means the system catches about 6 in 10 fraud circumstances. The conservative thresholds mirror a enterprise actuality: false positives create buyer friction, whereas missed fraud might be caught by different controls. Because the fraud sample database grows with confirmed circumstances, recall improves.
Value and velocity
The brand new answer diminished prices and processing time throughout each pipelines.
Part
Value discount
ID extraction (Amazon Textract + Amazon Rekognition + Claude)
91% discount vs. earlier answer
Fraud detection (Claude Sonnet 4 + Amazon Titan Embeddings + Amazon S3 Vectors)
3–5 seconds per picture
The ID extraction price represents a 91% discount from the earlier answer. This makes it economically viable to serve industries with lower-value microloans. The fraud detection pipeline completes in 3–5 seconds per picture.
Operational affect
Past accuracy and price, the answer modified how Solar Finance operates day-to-day:
- Handbook intervention projected to drop from 60% to 30% of functions, reducing the overview workload in half.
- Staffing projected to lower from roughly 3 FTEs to roughly 1 FTE for this {industry}.
- Area enlargement now economically viable for low-value mortgage economies.
- Adaptability—including a brand new doc kind or language requires immediate engineering and validation, not retraining specialised fashions.
Scalability and enlargement
The answer’s structure was designed for fast enlargement. Solar Finance operates throughout 9 international locations, and the serverless design permits industry-specific deployments with out infrastructure duplication. Including a brand new financial system requires configuration updates and redeployment. The staff updates Claude Sonnet 4 prompts through Amazon Bedrock and defines document-specific validation guidelines, then assessments in opposition to a validation dataset. These configuration modifications require redeploying the Lambda capabilities by the continual integration and steady supply (CI/CD) pipeline utilizing Terraform. The fraud detection system makes use of two complementary strategies. Visible sample detection through Claude Sonnet 4 identifies display pictures and digital manipulation. These methods are largely common throughout industries. Background similarity evaluation utilizing Amazon S3 Vectors catches fraud rings by evaluating backgrounds in opposition to identified patterns, with confirmed fraud circumstances added to enhance detection over time.
The modular structure permits steady enhancement. The AWS Step Capabilities orchestration permits including new fraud detection strategies as parallel Lambda capabilities with out disrupting present checks. These might be capabilities like EXIF metadata evaluation, gadget fingerprinting, and geolocation validation. Every would combine as further parallel checks with out requiring architectural modifications.
Classes realized
5 sensible takeaways from the engagement:
OCR + LLM beats LLM alone. Claude Sonnet 4 through Amazon Bedrock by itself achieved 61.8% accuracy for ID extraction, which was under the prevailing baseline. Including Amazon Textract for uncooked textual content extraction and utilizing Claude just for structuring jumped accuracy to 85%. The LLM is nice at understanding context and normalizing messy information. It’s not as dependable at exact character-by-character recognition from photographs.
Multi-tier OCR delivers resilience. The cascading strategy makes use of Amazon Textract as main and Amazon Rekognition as a fallback. No single OCR service dealt with each edge case, however the mixture added minimal price whereas serving to keep away from full failures on difficult photographs.
Fraud detection wants a number of strategies. Visible sample detection catches display pictures at 95%+ confidence. Background similarity catches fraud rings by location patterns. However background similarity solely achieves 55% recall on seen patterns and drops to 16.7% on novel patterns. Neither technique alone is ample, and the system improves as extra confirmed fraud circumstances are added to the database.
Begin easy, add complexity when metrics demand it. The staff achieved a 91% price discount by utilizing Amazon Textract as main OCR as a substitute of Claude for every little thing. They referred to as AnalyzeID solely when particular fields had been lacking and cached embeddings for fraud detection. Reserve costly fashions for duties the place they’re really wanted.
Serverless permits fast iteration. The parallel execution in AWS Step Capabilities lower fraud detection latency by 40% with minimal code modifications. The power to switch and deploy particular person Lambda capabilities with out downtime was essential throughout a 6-week engagement the place the strategy advanced weekly.
Subsequent steps
Solar Finance plans to construct on the proof-of-concept in a number of instructions.
- Develop visible detection. The present system solely checks for display pictures. It misses cartoons, illustrations, and AI-generated photographs. Increasing the detection immediate is the lowest-effort, highest-impact enchancment.
- Extra coaching information. Steady assortment of confirmed fraud circumstances and various background patterns will straight enhance background similarity recall past the present 55% on seen patterns.
- Further fraud indicators. Integrating EXIF metadata evaluation, gadget fingerprinting, and geolocation validation would add detection paths that don’t depend upon visible evaluation. That is significantly worthwhile for novel fraud patterns.
- Multi-language enlargement. Increasing to Solar Finance’s different economies in international locations throughout Southeast Asia, Africa, Latin America, and Europe requires language-specific immediate engineering and validation guidelines. Claude’s multilingual capabilities present a place to begin, and the staff is constructing a configuration framework to allow enlargement with out code modifications.
Clear up
For those who implement the same proof-of-concept, delete the next assets if you’re finished to keep away from ongoing prices:
- AWS Lambda capabilities created for the ID extraction and fraud detection pipelines.
- AWS Step Capabilities state machines.
- Amazon S3 buckets and Amazon S3 Vectors vector indexes used for fraud sample storage.
- Amazon API Gateway REST APIs.
- Amazon Cognito consumer swimming pools.
- AWS WAF internet entry management lists (ACLs).
- Any Amazon Bedrock provisioned throughput (if configured).
You possibly can delete these assets by the AWS Administration Console or by operating `terraform destroy` should you deployed the infrastructure utilizing Terraform.
Conclusion
On this submit, we confirmed how Solar Finance mixed Amazon Textract, Amazon Rekognition, and Amazon Bedrock to construct an AI-powered id verification pipeline. The answer improved extraction accuracy from 79.7% to 90.8%, lower per-document prices by 91%, and diminished processing time from as much as 20 hours to below 5 seconds. The core architectural sample, utilizing specialised OCR for textual content extraction and an LLM for clever structuring, applies to doc processing workflows the place conventional OCR falls quick. The serverless fraud detection system demonstrates how one can mix visible evaluation with vector similarity search to catch fraud patterns at scale.
For purchasers making use of for a microloan, that’s the distinction between ready a day and getting a solution whereas they’re nonetheless on their cellphone.
“Thanks to the AWS Generative AI Innovation Middle staff for an impressive partnership and really distinctive outcomes. What initially felt like an formidable — nearly unrealistic — goal has been reworked right into a safe, production-ready answer delivering measurable features in accuracy, velocity, and price effectivity. Particularly, the AI-powered fraud detection functionality — combining visible sample recognition and background similarity evaluation — represents a significant step ahead in defending our portfolio whereas sustaining a seamless buyer expertise. The affect on our operations and threat administration framework is fast and vital, and we deeply respect the experience, dedication, and execution excellence that made this doable.”
— Agris Vaselāns, Group CRO, Solar Finance
To learn the way generative AI can enhance your doc processing and fraud detection workflows, go to the Amazon Bedrock product web page or join with the AWS Generative AI Innovation Middle. For extra on OCR and doc processing, consult with the Amazon Textract Developer Information.
We’d love to listen to about your expertise with doc processing and fraud detection. Share your ideas within the feedback part.
In regards to the authors
Babs Khalidson is a Deep Studying Architect on the AWS Generative AI Innovation Centre in London, the place he makes a speciality of fine-tuning giant language fashions, constructing AI brokers, and mannequin deployment options. He has over 6 years of expertise in synthetic intelligence and machine studying throughout finance and cloud computing, with experience spanning from analysis to manufacturing deployment.
Vushesh Babu Adhikari is a Knowledge scientist on the AWS Generative AI Innovation middle in London with intensive experience in creating Gen AI options throughout various industries. He has over 7 years of expertise spanning throughout a various set of industries together with Finance , Telecom , Data Know-how with specialised experience in Machine studying & Synthetic Intelligence.
Luisa Bertoli is an AI Strategist on the AWS Generative AI Innovation Middle. She works with giant organizations on their AI technique, adoption, and multi-year transformation plans, serving to them transfer from experimentation to scalable, high-impact implementations. She has deep monetary providers area experience, constructed over years of designing and creating AI and ML merchandise within the {industry}.
Kimmo Isosomppi is a Senior Options Architect at AWS in Helsinki, Finland. He helps enterprise clients throughout the Nordic and Baltic areas flip complicated cloud and AI challenges into production-ready options, with specific experience in generative AI, agentic AI architectures, and cloud safety. He brings over twenty years of expertise throughout gaming, monetary providers, retail, and the general public sector.
Seppo Kalliomaki is an Account Govt at AWS in Tallinn, Estonia, specializing in enterprise cloud adoption and AI transformation throughout the Nordic and Baltic areas. Since 2017, he has helped organizations of their cloud journey and implement generative AI options, with specific experience in banking modernization, Public Sector providers, and rising AI use circumstances. Seppo works carefully with renewing cloud technique, migration planning, and AI adoption with AWS enterprise clients.
Nicolas Metallo is a Senior Deep Studying Architect on the AWS Generative AI Innovation Middle in Madrid. He designs and implements GenAI options utilizing Amazon Bedrock and SageMaker, together with fine-tuning LLMs, deploying multi-agent techniques, and main technical GTM for sovereign AI initiatives throughout EMEA.
Krišjānis Kočāns leads fraud prevention information science at Solar Finance Group throughout 14 international locations in 4 continents, constructing fraud detection techniques from scratch whereas driving Gen AI adoption.
Kaspars Magaznieks is Head of Fraud at Solar Finance – main Fraud prevention Crew, constructing fraud prevention framework, fraud prevention coverage. Kaspars has greater than 10 years’ expertise in fraud prevention working in world, quick paced lending firms!
Sergei Kiriasov is Head of Threat Know-how at Solar Finance, liable for shaping and delivering the expertise behind credit score threat decision-making. Main cross-functional collaboration between Threat and IT, ensures strong structure, environment friendly processes, and scalable options that empower information science, fraud prevention, and portfolio groups. With 15+ years in expertise, drives innovation and operational excellence throughout threat techniques.
