Ivan Plankin
Systems That
Never Go Down
I build AI infrastructure, Telegram automation, and high-performance backends in Rust. Production systems running for years without restarts. 10x lower infrastructure costs than Python/Node stacks.
struct TitanAgent {
pub id: Uuid,
pub capabilities: Vec<Skill>,
pub state: State<Active>,
}
impl TitanAgent {
fn orchestrate(&self) {
// Autonomous Execution Loop
self.execute()
}
}What I Solve
Production infrastructure that saves money and runs unattended.
AI Gateway & LLM Infrastructure
Unified API routing across OpenAI, Anthropic, Google. Automatic failover, cost optimization, usage analytics. 73K+ requests/month in production.
Telegram Automation & Analytics
AI-powered bots, channel analytics, automated content scheduling. Multi-account management, 24/7 autonomous operation. Anti-ban protection built-in.
High-Performance Rust Systems
Rewriting slow Python/Node services in Rust. 10x lower server costs, near-zero crash rate. Reproducible builds with Nix, declarative NixOS infrastructure, zero-downtime Colmena deployments.
Full-Stack Web Applications
SSR + WASM with Leptos. Islands architecture, zero JavaScript dependency. Nix flakes for reproducible builds, containerized deployment. 5-25ms response times.
How I Work
Structured methodology from discovery to production support.
Discovery
Understanding business goals, technical constraints, existing systems. Defining scope, success metrics, and risk boundaries before writing any code.
Architecture
System design, technology selection, risk assessment. Data flow diagrams, API contracts, infrastructure planning. Every decision documented.
Development
Test-driven development, continuous integration, iterative delivery. Code reviews, automated testing, performance benchmarks at every stage.
Deployment
Zero-downtime deployment, monitoring setup, production hardening. Container orchestration, health checks, automated rollback procedures.
Support
Post-launch optimization, performance monitoring, scaling guidance. Ongoing maintenance, security patches, architecture evolution as needs grow.
Architect, Not a Coder
I design systems that outlive frameworks. My code is infrastructure—built to scale, built to last.
AI-Assisted Development
Leveraging AI tools to accelerate development while maintaining architectural integrity. The human makes decisions; AI handles boilerplate.
Research & Development
Constantly exploring cutting-edge technologies. From WASM to distributed systems, I prototype tomorrow's solutions today.
Why Rust?
The same technology that powers Firefox, Dropbox, Cloudflare, and Discord. Here's what that means for your project:
| Metric | Rust | Traditional Stack | What It Means For You |
|---|---|---|---|
| Page Load Speed | 5-25ms | 200-800ms | Faster load = higher Google ranking + 7% more conversions per 100ms saved |
| Memory Usage | ~10MB | ~200MB | Lower hosting costs, handles 10x more users on same server |
| Security Vulnerabilities | ~0 | 70%+ CVEs | Microsoft/Google: 70% of CVEs are memory bugs. Rust prevents most of them. |
| Crash Rate | Near zero | Runtime errors | No null pointer exceptions, no "undefined is not a function" |
Bottom Line
Rust means your application is faster (better SEO, more conversions), cheaper to host (10x less server resources), and more secure. Predictable performance with no runtime surprises.
Rust-First Philosophy
Rust is my primary language and preferred tool for building reliable systems.
Its compile-time guarantees, unmatched performance, and full-stack capabilities provide a strategic advantage. Having one language from the database to the UI means one mental model and zero context switching.
While my focus is strictly on Rust for core infrastructure, I remain pragmatic and open to using other technologies when they are the right fit for the project.
Selected Works
Production systems built with Rust
HERMES_AI_PLATFORM
Telegram marketing automation for businesses. Schedules posts, manages multiple accounts, generates AI content. Runs 24/7 without manual work.
ANTIGRAVITY_SERVER
Unified API for multiple AI providers (OpenAI, Anthropic, Google). Automatic failover, cost optimization, usage analytics. NixOS deployment with Colmena.
NEXUS
AI video processing pipeline. Face tracking, smart cropping, multi-format adaptation. Automated content transformation for social media distribution.
TDATA-RS
Open-source library for parsing Telegram Desktop sessions. Secure encryption handling, published on crates.io. Used by developers worldwide.
DORODNYAYA_LANDING
IN DEVProfessional landing page for a medical specialist. Fast loading, SEO-optimized, mobile-first design. Deployed in containers.
GEMINI_PROXY_ROTATION
ARCHIVEDAPI proxy for Google Gemini with key rotation. Handles rate limits automatically, ensures uninterrupted service.
B2B_LEADGEN_ENGINE
LEARNING_STAGEDEPRECATED[DEPRECATED] B2B lead generation platform. AI conducts sales conversations and routes qualified leads. Legacy Python project.
Uplink
Direct channel for technical collaboration