Skip to content
BackNov 7
Sergey K.
Upwork badge

Sergey K.

Computer Vision & Mobile AI Ninja | YOLO, Edge AI, CoreML | LLM Agents

I build computer vision, mobile AI, and LLM agent systems that run in production — not proof-of-concepts. $600K+ earned · 100% Job Success · Top Rated Plus · 50+ apps shipped · 10+ years.   🎥 Computer Vision & Video AI Real-time detection, tracking, and video pipelines built for live production. - Built real-time tracking for a US sports company — processing live camera feeds frame-accurate for automated broadcast. - Image registration & feature matching (homography, SIFT/ORB) for frame alignment and multi-frame processing. - Edge & on-device optimization: quantization, ONNX / TFLite / CoreML export, low-latency real-time inference. Stack: YOLO, OpenCV, C++ inference, homography, SIFT/ORB, FFmpeg, ONNX, CoreML, TensorFlow Lite.   📱 Mobile AI — iOS & Android Where mobile meets ML: I ship AI that runs on the device, not just calls to a server. - On-device inference with CoreML & TensorFlow Lite — quantized, optimized models running in real time, offline, no backend round-trip. - Real-time computer vision on phones — live camera detection & tracking running directly on iOS/Android hardware. - AI embedded in the core product flow — vision and ML features, not a bolted-on toggle. - 50+ apps shipped over 10+ years across healthcare, fintech, wellness, and sports. Swift, SwiftUI, Kotlin, Jetpack Compose, Flutter, React Native.   📊 Data & ML Pipelines - Built a PySpark clustering pipeline for a US energy AI company — segmented ~1M households into 50–200 cohorts, production-grade with a validation framework. 🛠 Stack Vision: YOLO, OpenCV, C++ inference, homography, FFmpeg, CoreML, TensorFlow Lite Mobile: Swift, SwiftUI, Kotlin, Jetpack Compose, Flutter, React Native AI/LLM: Claude, GPT, Gemini, LangChain, LangGraph, RAG, vector DBs, n8n ML/Data: Python, PyTorch, TensorFlow, PySpark, Pandas, scikit-learn Cloud: AWS Bedrock, Azure OpenAI, Vertex AI, Firebase 🌎 Based in Buenos Aires (GMT-3) — solid overlap with US & EU business hours. English: Fluent. Weekly updates, clear timelines, honest scoping — if something needs to change mid-project, you hear it early, with options. Send me the problem — I'll tell you how I'd approach it and how long it'll take. Message me to start. :_: :D

🇦🇷Buenos Aires, Argentina$67/hr100% JSS4.87 (55)Since Jun 2014

Earnings and projects

MRR
$3.5k

Last 6 months ÷ 6

World#9,811 / 412k
🇦🇷Country#116 / 4,639
Revenue, last 30 days
$1.2k

Last 6 months: $21k

World#35,741 / 412k
🇦🇷Country#413 / 4,639
Total earnings
$660k

Billed in all: $671k

World#1,516 / 412k
🇦🇷Country#11 / 4,639
Per project
$9.2k

$3k in the last 6 months

World#17,728 / 412k
🇦🇷Country#216 / 4,639
Projects, last 6 months
7

Across 8 contracts

75% fixed25% hourly
World#28,826 / 412k
🇦🇷Country#288 / 4,639
Projects in all
72

Across 86 contracts

33% fixed67% hourly
World#41,474 / 412k
🇦🇷Country#368 / 4,639

Earnings, last 30 days

$671k+$1.2k earned in the last 30 daysBilled in all, before Upwork's fee · drag across the chart to see what was earned in any stretch
$669k$670k$671k$672kOct 8Oct 23Nov 7

Skill rankings: Top 3% in Object Detection

Ranked by MRR among everyone on Upwork with the skill.

  1. 2Wear OS18 freelancers · 🇦🇷 #1Top 11%
  2. 3Image Recognition77 freelancers · 🇦🇷 #1Top 4%
  3. 3TensorFlow Lite23 freelancers · 🇦🇷 #1Top 13%
  4. #4Edge AI64 freelancers · 🇦🇷 #1Top 6%
  5. #5Core ML61 freelancers · 🇦🇷 #1Top 8%
  6. #6Object Detection188 freelancers · 🇦🇷 #1Top 3%
  7. #8YOLO229 freelancers · 🇦🇷 #1Top 3%
  8. #59Smartphone995 freelancers · 🇦🇷 #1Top 6%