Who Is Emre Dikmen?

Short definition

Emre Dikmen is the founder and CEO of SilverOps, an İstanbul-based AI automation and custom software agency. He is a full-stack engineer in the full sense of the term: not only front-end and back-end, but database, servers, networking, cloud infrastructure, CI/CD pipelines, security, containers, CDN, monitoring and logging, and backup and recovery, every layer of the stack. He writes at different levels, from low-level systems languages to the web and mobile interface.

He treats the design, software, infrastructure, and AI sides not as separate parts but as a single system.

Technical foundation

The first language he learned, and the one he uses most, is C#. He also works in C, C++, Java, Python, Go, JavaScript, TypeScript, SQL, and Bash, and drops down to low-level languages like Assembly when needed, developing across layers from hardware-near systems code to a modern web and mobile interface.

His production stack is Next.js, React, TypeScript (strict mode, without any), Three.js / React Three Fiber, and GSAP. On the back-end and infrastructure side: Supabase with mandatory Row Level Security, PostgreSQL, Docker, nginx, and VPS.

He does not just write the systems; he also sets up and runs the deployment, CI/CD pipelines, containers, CDN, networking, monitoring and logging, and backup and recovery layers himself. As one person he covers the whole chain, from the front-end to operating the live system.

What he builds: production systems

His work can be illustrated by a few systems in production use:

• Nova: A JARVIS-style personal AI assistant with functions such as calls, messages, and status reports. An ongoing project.
• A full ERP system: An ERP system for an industrial company: 21 tables and 20 modules, an AI assistant working with live database context, four user roles, and layered authorization. In field use after pre-deployment hardening.
• A WhatsApp AI bot: A bot on the WhatsApp Business API with conversation memory, automatic lead detection, and owner notification. Built end to end, including the Meta Business Verification process.
• A fine-tuned model and security: A brand-specific fine-tuned language model project, and security hardening across multiple client systems.

An example of delivery speed: a 29-page corporate website in three languages, with AI-generated visuals, was built and deployed in under two hours.

Security

On the security side he has experience in pen-testing and exploit development, working from both an offensive and a defensive perspective. In practice this covers rate limiting, CSP, user enumeration protection, SQL injection, and exploit scripting.

He evaluates his own systems from an attacker's perspective and designs the defense accordingly, which carries directly into the security of the systems he builds.

Research and physics

Outside software, he has an interest in physics. He studied quantum physics and astrophysics on his own and wrote a single-author astrophysics paper: The Radius of the Heaviest Pulsar Excludes Observable Mass Twins in a Causal Layered Hybrid-Star Model.

The paper examines whether the mass twins phenomenon arising from a phase transition in dense matter can be consistent with current observational data, using a layered model that integrates the TOV and tidal equations, and concludes with a falsifiable prediction against future measurements. The subject falls within computational astrophysics.

Achievements, certifications, and community

He was world runner-up in a university-level robotics and mechatronics competition and holds the International Master title in chess.

He holds 17 Anthropic certifications, some of which cover the deployment of Claude on Amazon Bedrock and Google Vertex AI (these certifications are issued by Anthropic and cover deployment on those platforms; they are not issued by AWS or Google). He also holds 8 Meta Blueprint certifications and is a member of the Anthropic Claude Partner Network. He is part of the Google Developer Group (GDG) community in İzmir and İstanbul and a member of the OpenAI developer community.

How he works

What distinguishes the systems SilverOps builds from off-the-shelf bot services is the way of working. The system is built for the client (done-for-you) and integrated into existing software; WhatsApp and the voice channel are managed as a single system, often multilingually; when needed, data is kept on the client's own infrastructure and tool access is constrained with the Model Context Protocol (MCP).

Working with him

The process starts with a discovery call covering the business's workflow, the messages and calls that can be automated, the systems to connect, and the language needs. Pricing is fixed and package-based; the right package is confirmed in a short call and the price is transparent.

Frequently Asked Questions