Argo Saakyan - Principal Computer Vision Engineer

Summary

Building real-time perception for Physical AI at Agnify (Austin, TX) - a GPU-resident TensorRT pipeline serving 14 live CCTV cameras (~400M frames/month), benchmarked to 350 fps on a single GPU.
Main author of D-FINE-seg - open-source detection / instance / semantic segmentation framework (200+ GitHub stars, arXiv paper, pip install dfine-seg) that beats YOLO26 and RF-DETR on Cityscapes with 2–3× fewer parameters - now the core of Agnify's perception stack.
Led fraud detection at Veryfi (YC17) as tech lead and core owner - built and deployed 13 vision models processing real documents, each handling 5–12 million images per month.
Optimized-inference specialist: TensorRT, OpenVINO, ONNX, CoreML, Triton - ~2 ms end-to-end detection on a single consumer GPU, down to 700 KB image classifiers on-device.
Own my product area end-to-end and work directly with the CEO/CTO - the last 5 companies were startups, where I also led ML engineers and data/annotation workflows.
Speaking at conferences (DevDays Europe, Conf42, Warsaw IT Days), writing about CV/DL (arXiv, Towards AI), judging international awards.

                        6
                        Years Experience
                    

                        20+
                        Models Shipped
                    

                        200+
                        GitHub Stars
                    

                        400M
                        Frames / Month
                    

Experience

Principal Computer Vision Engineer

Agnify - Austin, Texas

Apr 2026 - Present

Physical AI startup giving factories real-time visual understanding of their production lines.

Built the core perception stack as author of D-FINE-seg - object detection, instance / semantic segmentation.
Engineered a fully GPU-resident TensorRT pipeline (NVDEC → preprocess → inference → postprocess → NVENC) serving 14 live CCTV cameras (~400M frames/month) - benchmarked to 350 fps on a single GPU, capped at 10 fps/camera to free compute for downstream tasks.
Own all model integrations across the stack, from data mining to serving - full train → export → bench → infer pipelines (TensorRT / OpenVINO, cloud and edge) for D-FINE-seg, X3D video, timm classifiers, and V-JEPA2; inference-only integrations for ViT-Pose, SAM3, and Qwen/Gemini VLMs.
Built features: object tracking, counting, pose estimation, and a SAM3 auto-labeling solution.
Achieved 99.8% accuracy on a bottle-counting task (D-FINE-seg → tracker → counter).

PyTorchTensorRTOpenVINOSAM3V-JEPA2ByteTrackQwen (VLM)vLLMNVDEC/NVENCDockerLinux

Previously

Lead Computer Vision Engineer

Veryfi (YC17) - Silicon Valley

Jun 2023 - May 2026

Tech Lead of the Fraud Detection suite; built the decision engine fusing vision + text model signals into a final fraud score.

Led a team of fraud-detection ML engineers and coordinated delivery across teams.
Built and deployed 13 production vision models (classification, detection, segmentation), each processing 5–12M images/month.
Designed custom conv/transformer architectures beating prior models on both F1 and latency (raised one from 93% to 99%, F1 0.98).
Shipped 700KB on-device models (limit 1.5MB) at 7fps; ONNX / TorchScript / TensorRT / TFLite backends.

PyTorchTensorRTTorchScriptTFLiteONNX

CV Engineer & Adviser Retinal pathology segmentation, real-time billiards analysis (60fps), factory defect detection

CV Researcher - Diagnocat 3D dental imaging, medical AI deployed across multiple countries

CV Engineer - DisArm Gun detection, real-time video surveillance, edge deployment

DL Engineer - BigDataMSU Infrastructure monitoring, industrial anomaly detection, hardware prototyping

Data Scientist / Analyst Healthcare analytics, insurance, forecasting & automation

Founder - Earthome IoT smart home systems, hardware prototyping, 10+ deployments

Open Source & Projects

Featured

D-FINE-seg

Object Detection & Instance Segmentation Framework

Real-time object detection, instance and semantic segmentation in one framework - segmentation heads implemented from scratch on D-FINE, not a fork. Full pipeline: dataset tools → training (DDP, EMA, mosaic) → export (ONNX, TensorRT, OpenVINO, CoreML, LiteRT) → benchmarked multi-backend inference. Beats YOLO26 and RF-DETR on Cityscapes with 2–3× fewer parameters at ~2 ms end-to-end. Published with an arXiv paper, installable via pip install dfine-seg, and now the core of Agnify's production stack - 14 live cameras, 99.8% counting accuracy.

GitHub Paper

PyTorchObject DetectionSegmentationONNXTensorRT

Triton Server Pipeline

Open-source inference pipeline built on NVIDIA Triton Inference Server for scalable, optimized model serving in production environments.

GitHub

TritonTensorRTDockergRPC

PyTorch Training Pipeline

Reusable, modular training pipeline for computer vision tasks - designed for rapid experimentation and clean reproducibility.

GitHub

PyTorchHydraAlbumentationstimm

Hardware & Prototyping

Physical prototypes with microcontrollers, single-board computers, and sensors (lidar, radar, ultrasonic). Linux workstation builds. Woodworking & metalworking.

Nvidia JetsonRaspberry PiArduinoSensorsLinux

Technical Skills

Deep Learning & CV

PyTorch TorchVision timm / HuggingFace OpenCV Albumentations ByteTrack SAM 3 VLMs (Qwen, Gemini) Lightning AI NumPy Scikit-Learn

Optimization & Deployment

TensorRT OpenVINO ONNX CoreML TensorFlow Lite TorchScript Triton Inference Server NVDEC / NVENC

Infrastructure & Tools

Linux Docker Git Flask FastAPI Hydra Python

Hardware & Prototyping

Linux Workstations Nvidia Jetson Microcontrollers Single-Board Computers Soldering Lidar / Radar / Ultrasonic Woodworking Metalworking

Domains

Object Detection Instance Segmentation 3D Segmentation Classification Video / Action Recognition Object Tracking Pose Estimation Counting Anomaly Detection Medical Imaging Edge Inference

Speaking, Writing & Awards

Conferences

DevDays Europe 2023/24 Conf42 ML 2023 Voxel51 Warsaw IT Days 2023

Published in

ArXiv Towards Data Science Towards AI DevGenius

Awards

CODiE Award Judge Hackathon Judge Digital Breakthrough Winner

Let's Connect

Interested in working together, have a project in mind, or just want to chat about computer vision, robotics, or edge AI? I'd love to hear from you.

LinkedIn [email protected] GitHub