AI / ML Engineer • Computer Vision • Generative AI
Building practical AI systems, one focused iteration at a time.
Building intelligent systems that see, understand, and create.
I am a Machine Learning Engineer focused on Computer Vision and Agentic AI, with a strong foundation in scalable backend systems. My engineering philosophy revolves around translating complex research papers into optimized, production-ready code.
- 🎯 Focus: Bypassing computational bottlenecks in high-resolution (4K) object detection using Explainable AI (XAI).
- 🤖 AI Engineering: Building local LLM agents that seamlessly interact with third-party ecosystems (Google APIs, etc.).
- ⚙️ Infrastructure: Architecting robust database migrations and building backend profilers.
- 💡 Goal: I build systems that are not just intelligent, but fast, scalable, and resilient.
A sleek, dark-themed control panel designed for decoupled ML microservices and robust task queues, eliminating UX bottlenecks with pure speed and instantaneous rendering. Key Innovations:
|
A novel coarse-to-fine computer vision pipeline designed for efficient small object detection in high-resolution (2K/4K) aerial imagery. Tackles the critical trade-off between resolution and latency in drone forensics. Key Innovations:
|
A fast neural style transfer implementation that generates stylized images using a feed-forward CNN trained with perceptual loss. Performs instant stylization in a single forward pass. Key Features:
|
|
🧭 pygog (Google CLI Agent) |
📐 Depth Estimation + Semantic Seg. |
I actively contribute to the broader developer ecosystem, focusing heavily on AI tooling, backend infrastructure, and application security:
- 🤖 SynapseKit/SynapseKit: Led additions of multiple LLM providers and vector retrieval backends to broaden model support and storage options. Also built Discord automations for onboarding, moderation, and docs workflows to improve community operations.
- 🧠 pydantic/pydantic-ai: Upgraded the Anthropic code‑execution tool integration to align with newer API versions. This keeps model tooling stable as providers evolve.
- 🧪 matorral-project/matorral: Wrote admin‑focused backend tests, including coverage for
make_superuser. That hardens privileged flows and reduces regressions in management commands. - 🔒 trusera/ai-bom: Contributed to AI SBOM generation that audits project workflows. Outputs CycloneDX‑compliant dependency reports for stronger supply‑chain security.
- ⚡ Nikolaev3Artem/fastapi-silk: Implemented SQLite + Alembic migrations to establish durable database setup. Expanded profiler test coverage to improve reliability and maintenance.