Alexander Myasoedov
Alexander Myasoedov is an individual software engineer, machine-learning enthusiast, open-source developer, founder, and technical writer—not an organization. His public work focuses on production software, LLM applications and security, AI evaluation, and developer tooling.
At a Glance
- Independent software developers and AI researchers
- Developers building LLM applications, agents, and LangChain services
- AI security and red-teaming teams
- Solo founders and startups shipping AI-enabled applications
- +1 more
AI Tools by Alexander Myasoedov
(1)
Agentic Security
Open Source LLM Vulnerability Scanner
Discussions
No discussions yet
Be the first to start a discussion about Alexander Myasoedov
Latest News
Published 'Should You Ban AI-Generated Code? AI Policy 2026' on Metaheuristic.
Published 'JEV: Tool Calling in Production' on Metaheuristic.
Published 'The Slop Grenade: AI Code' on Metaheuristic.
Published 'How Much Does It Cost to Build an AI Agent?' on Metaheuristic.
Products & Services
Open-source Agentic LLM Vulnerability Scanner / AI red-teaming kit, distributed as the Python package agentic_security under Apache License 2.0. It tests agent workflows and LLMs for jailbreaks, multimodal attacks, and other weaknesses.
A developer tool for automatically serving LangChain LLM applications and agents with FastAPI; its repository describes it as LLMOps and supports token tracking, conversational retrieval, custom functions, and deployment-oriented workflows.
A live-application security evaluation service associated with Myasoedov as founder. It scans deployed applications, looks for access-control gaps and exposed secrets, produces evidence-backed findings, and provides human security-engineer verification and remediation guidance.
A current technical publishing project by Alexander Myasoedov covering agents, retrieval, evaluations, AI-generated code, tool calling, and operational guardrails.
Market Position
Myasoedov's portfolio combines an Apache-licensed, self-hostable LLM/agent vulnerability scanner with developer-oriented LLM serving and a commercial live-application verification service. Agentic Security emphasizes customizable and adaptive attacks, multimodal testing, fuzzing, and API stress testing; VibeEval differentiates around deployed-app testing, reproducible evidence, and human verification rather than a purely automated scan.
Leadership
Founders
Alexander Myasoedov
Software engineer and machine-learning enthusiast; GitHub describes him as a software engineer, FP and machine-learning enthusiast. His public profile describes technical leadership, infrastructure and architecture work, and experience with Kafka, Cassandra, distributed caching, Mesos/Marathon/Kubernetes, video-streaming analytics, Python, Go, Docker and AWS. He is listed as Founder of VibeEval and has a bachelor's degree in Physics from National Technical University of Ukraine 'Kyiv Polytechnic Institute' (2012).
Founding Story
No separate corporate founding story is published for Alexander's personal work. His project portfolio shows a progression from software/infrastructure and developer tooling into LangChain/LLM serving, AI red-teaming, and live-application security evaluation.
Business Model
Revenue Model
The open-source Agentic Security and Langcorn projects are distributed as developer software rather than a disclosed paid service. VibeEval uses a subscription and one-time-purchase model for live-app security scans, with a 14-day free trial.
Pricing Tiers
For solo builders; unlimited projects, engineer-verified findings, daily re-scans, and 24-hour email support.
Five seats; scheduled scans, shared reports, GDPR/SOC2/HIPAA gap analysis, MCP and webhook API, and priority support.
Pro coverage with no renewal fees, plus real-time monitoring and a 30-day money-back guarantee.
Target Markets
- Independent software developers and AI researchers
- Developers building LLM applications, agents, and LangChain services
- AI security and red-teaming teams
- Solo founders and startups shipping AI-enabled applications
- Growing engineering teams needing shared security coverage and compliance gap analysis
- Developers and researchers red-teaming LLMs and agent workflows
- Security teams testing jailbreak, prompt, multimodal, and API vulnerabilities
- CI checks and regression testing for AI safety defenses
- Serving and operating LangChain applications and agents with FastAPI
- Founders and startup teams checking deployed applications before launch
- Testing access-control boundaries and cross-tenant data exposure