Mert Palazoğlu
Mert Palazoglu has been an industry analyst at AIMultiple since 2023.
Research interests
His work focuses on the latest trends, companies, and innovations in:- Artificial intelligence (AI)
- Cyber security with a focus on network security
- Customer service software
- Proxies for data collection
Education
He graduated with a BS in Management from Bilkent University in 2021.
Latest Articles from Mert
15 Security Threats to LLM Agents (with Real-World Examples)
Even a few years ago, the unpredictability of large language models (LLMs) would have posed serious challenges. One notable early case involved ChatGPT’s search tool: researchers found that webpages designed with hidden instructions (e.g., embedded prompt-injection text) could reliably cause the tool to produce biased, misleading outputs, despite the presence of contrary information.
Top PAM Solutions: 8 Commercial Vendors + Free Alternatives
We spent three days testing and reviewing popular Privileged Access Management (PAM) solutions. We used the free trials and admin consoles of BeyondTrust, Keeper PAM, and ManageEngine PAM360. For solutions that required registration, we relied on official product documentation and user experiences to assess their capabilities.
The LLM Evaluation Landscape: 16 Frameworks by Functionality
We spent 2 days reviewing popular LLM evaluation frameworks that provide structured metrics, logs, and traces to identify how and when a model deviates from expected behavior.
Building Personal AI Agents + 18 Agent Platforms and Tools
We spent the two days experimenting with real-world demos and tools to build personal AI assistants that can handle your tasks, such as scheduling meetings, managing notes, or sorting through emails. We will dive into three main approaches to building and using personal AI assistants, with real-world examples for each: 1.
Building AI Agents with Anthropic's 6 Composable Patterns
We spent 3 days experimenting workflows and agent pipelines in n8n according to Anthropic’s and OpenAI’s guides on building effective AI agents. We are going to distill down everything we have learned to give you a guide to build functional AI agents in your LLM projects.
Low/No-Code AI Agent Builders: n8n, AgentKit, make, Zapier
We spent three days setting up and configuring AI agent workflows using the free tiers of popular low/no-code tools, including n8n (self-hosted), make, and Zapier, and evaluated OpenAI’s AgentKit based on its official documentation.
Best 7 AI Testing Platforms for QA
We evaluated AI testing platforms embedded with AI agents; most were overhyped Selenium/Playwright with marketing. A few were capable of writing/maintaining test cases or visual testing, though even these tools still have notable limitations. From these, we selected 7 platforms and categorized them by their primary focus areas.
LCMs: From LLM Tokenization to Concept-level Representation
Large concept models (LCMs), as introduced by Meta in their work on “Large Concept Models,” represent a fundamental shift away from token-based prediction toward concept-level representation.
Authorization for AI Agents: Permit.io, Descope & more
I have been exploring agent identity and the authentication/authorization platforms that could support it, while also examining how standards like OAuth 2.0 and frameworks such as Keycloak might apply. Below, I listed the best AI agent–specific platforms and features, categorized by their primary focus.
How we Moved from LLM Scorers to Agentic Evals?
Evaluating LLM applications primarily focuses on testing an application end-to-end to ensure it performs consistently and reliably. We previously covered traditional text-based LLM evaluation methods like BLEU or ROUGE. Those classical reference-based NLP metrics are useful for tasks such as translation or summarization, where the goal is simply to match a reference output.
AIMultiple Newsletter
1 free email per week with the latest B2B tech news & expert insights to accelerate your enterprise.