BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//events//Events//EN
CALSCALE:GREGORIAN
X-WR-CALNAME:Oct 15 - AI\, ML\, and Computer Vision Meetup
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
BEGIN:VEVENT
UID:event-58rpb3u4@ontown.app
DTSTAMP:20261008T195834Z
DTSTART:20261015T160000Z
SUMMARY:Oct 15 - AI\, ML\, and Computer Vision Meetup
LOCATION:San Francisco\, United States
DESCRIPTION:AI Professionals San Francisco Machine Learning & Agentic AI\nJ
 oin our virtual meetup to hear talks from experts on cutting-edge topics a
 cross AI\, ML\, and computer vision.\n\nTime\, Place and Location\n\nOct 1
 5\, 2026\n9:00 AM - 11:00 AM PST\nOnline. Register for the Zoom! (https://
 voxel51.com/events/ai-ml-and-computer-vision-meetup-october-15-2026)\n\nTe
 sting AI Systems in Production: Data Quality\, Drift\, and Model Evaluatio
 n\n\nAI systems can pass offline evaluation and still fail in production w
 hen real-world data changes\, features become stale\, labels or feedback s
 ignals are incomplete\, or model behavior drifts away from expected outcom
 es. This talk shares practical patterns for testing and evaluating AI syst
 ems after deployment\, including data quality checks\, drift detection\, o
 nline/offline metric comparison\, model monitoring\, and rollback analysis
 .\n\nUsing personalization and recommendation systems as examples\, we wil
 l examine how teams can build evaluation workflows that catch quality issu
 es before users do. Attendees will leave with a practical checklist for ma
 king AI-backed systems easier to evaluate\, debug\, and operate as data ch
 anges over time.\n\nAbout the Speaker\n\nJayakumar Ramalingam (https://www
 .linkedin.com/in/jayakumarramalingam) is a Staff Software Engineer and Clo
 ud Architect at SiriusXM with over 16 years of experience building cloud-n
 ative platforms\, real-time data pipelines\, resilient APIs\, and AI/ML-en
 abled applications at production scale.\n\nWhere Should Your Model Live? A
  Framework for Tiering Computer Vision Deployments\n\nWhere should a compu
 ter vision model actually run - on-device\, near the edge\, or in the clou
 d? It's a decision that looks simple until requirements like latency\, cos
 t\, connectivity\, and update cadence start pulling in different direction
 s\, often revealing themselves only after deployment.\n\nDrawing on hands-
 on experience developing and deploying CV models across Hailo\, Nvidia\, Q
 ualcomm and AWS platforms\, this talk introduces a practical framework for
  tiering computer vision deployments based on real project requirements an
 d constraints. Discussion will include what changes at each tier - from de
 velopment to deployment to monitoring and update strategy - with relevant 
 industry examples.\n\nAttendees will leave with a set of questions or a fr
 amework they can use to place their own CV projects into the right tier.\n
 \nAbout the Speaker\n\nAjaykumaar Sivacoumare (https://www.linkedin.com/in
 /ajay-sa/) is an AI Software Engineer specializing in computer vision and 
 edge AI\, with production experience developing and deploying CV models ac
 ross Nvidia\, Hailo\, Qualcomm and AWS-based platforms.\n\nFrom 2D Slices 
 to 3D Tumors: Lightweight Volumetric Detection Without Heavy 3D Networks\n
 \nSlice-wise 2D detectors are fast and scalable\, but they struggle to pro
 duce reliable 3D bounding boxes from volumetric medical data. This talk pr
 esents YOLO-PVC\, a lightweight post-processing framework that consolidate
 s slice-wise YOLO detections into coherent 3D bounding boxes using percent
 ile-based geometric aggregation and a minimal MLP calibration module.\n\nT
 his talk demonstrates consistent improvements in volumetric IoU across thr
 ee liver tumor categories i.e.\, HCC\, CCA\, and Mixed\, without requiring
  dense 3D annotations or memory-intensive architectures. The talk covers t
 he clinical motivation\, the technical approach\, and practical lessons fr
 om deploying computer vision on real hospital MRI data.\n\nAbout the Speak
 er\n\nTalha Waqas (https://www.linkedin.com/in/mtalha-waqas-255557144/) is
  a second-year PhD student at ESME Research Lab\, Paris and LISSI\, Univer
 sité Paris-Est\, working on computer vision applied to medical imaging\, 
 with a focus on tumor classification\, detection\, and segmentation in mul
 ti-phase liver MRI.\n\nThe Two-Loop Architecture for Voice AI\n\nBuilding 
 responsive voice AI requires balancing latency with intelligence. This tal
 k introduces a practical architecture that separates real-time conversatio
 n from asynchronous reasoning\, enabling richer interactions without slowi
 ng the user experience. The session covers reusable design patterns drawn 
 from production-inspired conversational AI systems.\n\nAbout the Speaker\n
 \nAbhinav Tushar (https://lepisma.xyz/wiki/about/) is an ML engineer and r
 esearcher specializing in Conversational AI and Speech Technology.
URL:https://ontown.app/e/58rpb3u4-oct-15-ai-ml-and-computer-vision-meetup/
END:VEVENT
END:VCALENDAR