Services
Contact Us
Şevval Alper

Şevval Alper

AI Researcher
19 Articles
Stay up-to-date on B2B Tech
Şevval is an AI researcher at AIMultiple. She has previous research experience in pseudorandom number generation using chaotic systems.

Research interests

Şevval focuses on AI coding tools, AI agents, and quantum technologies.

She is part of the AIMultiple benchmark team, conducting assessments and providing insights to help readers understand various emerging technologies and their applications.

Professional experience

She contributed to organizing and guiding participants in three “CERN International Masterclasses - hands-on particle physics” events in Türkiye, working alongside faculty to facilitate learning.

Education

Şevval holds a Bachelor's degree in Physics from Middle East Technical University.

Latest Articles from Şevval

AI
Benchmark
Aug 14

AI Code Review Tools Benchmark

With the increased use of AI coding tools, codebases have become more prone to vulnerabilities, which increased the need for effective code reviews. To address this, we introduce RevEval (AI Code Review Eval), which benchmarks the top four AI code review tools across 309 pull requests from repositories of varying sizes and evaluates their performance…

Agentic AI
Benchmark
Aug 14

AI Agent Platforms Benchmark: Claude Managed Agents vs Google Vertex Agent Engine

We benchmarked 4 AI agent platforms across 3 dimensions: task completion (10 coding tasks × 3 runs), harness-specific capabilities (steering, reconnection, long-conversation recall, large-file handling), and cost. Claude Managed Agents and Vertex AI Agent Engine both achieve 100% pass rates on the task suite, with Vertex winning on cost ($1.45 vs $2.50). For harness-specific features…

Agentic AI
Benchmark
Aug 14

Code Execution with MCP: A New Approach to AI Agent Efficiency

Anthropic introduced a method in which AI agents interact with Model Context Protocol (MCP) servers by writing executable code rather than making direct calls to tools. The agent treats tools as files on a computer, finds what it needs, and uses them directly with code, so intermediate data doesn’t have to pass through the model’s…

AI
Benchmark
Aug 12

OCR Benchmark: Text Extraction / Capture Accuracy

OCR accuracy is critical for many document processing tasks, and SOTA multi-modal LLMs are now offering an alternative to OCR. We benchmarked leading OCR services in DeltOCR Bench to identify their accuracy levels in different document types: The full names of the above products and their versions in use as of November 2025 are listed…

AI
Benchmark
Aug 12

Best AI Code Editor: Cursor vs Windsurf vs Replit

Making an app without coding skills is highly trending right now. But can these tools successfully build and deploy an app? We benchmarked 6 AI code editors across 10 real-world web development challenges. Each task required implementations such as backend, frontend, authentication, state management. We evaluated backend correctness, frontend behavior, and combined performance, and analyzed…

Agentic AI
Benchmark
Aug 11

Top Agent Harnesses: Claude Code vs Codex

Agent harnesses serve as the production runtime for AI agents, with design choices that create performance variation across identical underlying models. We benchmarked 17 agent harnesses across 10 coding tasks. To isolate the harness rather than the model, we ran every agentic CLI on a single foundation model, Claude Sonnet 4.6 (non-reasoning), and AI code…

AI
Benchmark
Aug 11

Agentic IT: Can LLMs Design a Benchmark

We gave 12 large language models the job a benchmark team does: invent a benchmark, build it, run four models through it, and report the results. Each did it twice. None of the 24 attempts passed every criterion, and six of the rubric’s checks were passed by none of them. The two topics are text-to-SQL,…

Agentic AI
Benchmark
Aug 10

VELC-Bench: Verification on Long Context Benchmark

The model’s ability to locate a specific metric in context, compare its value to a claim, and confirm or reject it. This tests fine-grained value matching under long-context conditions. The model must both retrieve the value and perform a precise comparison. The models are tested in the following context windows: claude-fable-5 scores 90.0% on verify…

AI
Benchmark
Aug 4

Screenshot to Code: Lovable vs v0 vs Bolt

During my 20 years as a software developer, I led many front-end teams in developing pages based on designs that were inspired by screenshots. Designs can be transferred to code using AI tools. While expecting a pixel-perfect transfer is wrong in the current state of the tools, they can give developers a foundation to work…

AI
Benchmark
Aug 4

Audience Simulation: Can LLMs Predict Human Behavior?

In marketing, evaluating how accurately LLMs predict human behavior is crucial for assessing their effectiveness in anticipating audience needs and recognizing the risks of misalignment, ineffective communication, or unintended influence. Audience simulation with LLMs enables the modeling of virtual audiences, helping organizations anticipate reactions to content or products without relying on costly surveys or focus…