Skip to content
View swd07's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report swd07

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
swd07/README.md
Eduard Kharaev — Applied AI Engineer and AI Solutions Architect

Typing SVG


Portfolio Telegram Email


Remote Contract Relocation


Open to Senior Applied AI · Forward-Deployed Engineering · AI Solutions Architecture roles



Production · Live system · Engineering · Stack · Principles · Repositories


⚡ Production AI, end to end

I design, build and operate production AI systems with end-to-end technical ownership — from computer-vision models and GPU inference to LLM agents, business integrations, backend/data layers and product interfaces.

My bias is toward measurable impact, honest evaluation and systems that stay reliable after the demo is over.

~3,800

Retail outlets
~1,500 active monthly

9–11k

Orders / month
98% via platform

0.91

Detection F1
on unseen shelf images

95.8%

Brand precision
confirmed end-to-end evaluation
🧠 20+ typed tools
production agent layer
⚡ NVIDIA H200
self-hosted AI inference
🤖 Self-hosted Qwen
Qwen2.5-VL 72B · Qwen3.6 35B
served with vLLM
🎯 73.1% SKU precision
end-to-end retrieval + fusion

🔒 Most production code is private because it runs inside commercial systems with proprietary data and business integrations. The public repositories below contain sanitized case studies, architecture, metrics, evaluation methodology and runnable examples.

👁 What production looks like

Live retail shelf detection output

Real shelf photo through the production CV pipeline — detections, SKU labels, price-tag reads and explicit Unknown when evidence is insufficient.

🧩 One system, not isolated models

flowchart LR
    A[Retail shelf photo] --> B[GroundingDINO detection]
    B --> C[Product crops]
    C --> D[Qwen2.5-VL OCR / package evidence]
    D --> E[Qwen3-Embedding-8B]
    E --> F[Qdrant dense retrieval]
    C --> G[DINOv2 visual k-NN]
    C --> H[ArcFace metric retrieval]
    F --> I[Deterministic fusion + guardrails]
    G --> I
    H --> I
    D --> I
    I --> J{Enough evidence?}
    J -->|Yes| K[SKU / brand match]
    J -->|No| L[Unknown / abstain]
Loading

Production principle: confidence is not correctness. A real AI system should be able to abstain, run in shadow before promotion, be evaluated on real populations and roll back cleanly.

🚀 Featured engineering

Production case studies with architecture, metrics, engineering decisions and evaluation methodology.

AI Platform Agents AIOps Voice AI Product Engineering

Includes:

  • FMCG commercial operating platform
  • infrastructure monitoring agent
  • real-time voice assistant
  • coaching / fitness product

Production-engineered retrieval + computer-vision merchandising pipeline with explicit uncertainty.

Detection OCR VLM Qdrant DINOv2 ArcFace

Evidence:

  • 95.8% brand precision
  • 73.1% SKU precision
  • 47k-box replay harness
  • shadow → active rollout gates

Multi-source marketing intelligence platform connecting social, website, search and campaign analytics.

Instagram API Search Traffic Content Intelligence AI Reporting

Measured window:

  • +2,859 Instagram followers
  • +88% website visits
  • 17.6% Google Search CTR
  • 2.4 average search position

🛠 AI / ML stack

Core engineering stack



Qwen Hugging Face vLLM Qdrant OpenCV Langfuse MCP Prometheus Grafana

Computer Vision · OCR · Vision-Language Models · Dense Retrieval · Retrieval-Augmented Systems · Vector Search · LLM Agents · Tool Calling · MCP · GPU Inference · Evaluation · Observability · Marketing Intelligence

🧬 Engineering DNA

🎯 Evaluate first

Golden-set regression, replay, acceptance gates and population-level validation before promotion.
🛡 Safe rollout

Shadow → measure → gate → active, with explicit rollback paths.
⚙️ Deterministic where possible

LLMs augment reliable systems; they do not replace reliable logic without reason.
🏗 End-to-end ownership

Model layer, backend, data, integrations, frontend, deployment and operations.

📦 Repositories & recent work

AI Platform Portfolio Retail Shelf Detection Marketing Platform



GitHub Profile Portfolio


Building AI systems that actually reach production.

Applied AI · Production Retrieval · Computer Vision · LLM Systems · AI Infrastructure · Growth Systems · Product Engineering


Portfolio Telegram

Popular repositories Loading

  1. ai-platform-portfolio ai-platform-portfolio Public

    Production AI systems end-to-end: LLM agents, CV pipelines, self-hosted GPU inference. Case studies with architectures and metrics

  2. swd07 swd07 Public

  3. retail-shelf-detection retail-shelf-detection Public

    Production CV pipeline for retail shelf photos: share-of-shelf, SKU recognition, price tags. Case study with architecture, guardrails and honest metrics (F1 0.68→0.91).