Kargo

Come join us! Kargo Chicago Advertising Federation

Great to spend the week at the Kargo NYC HQ with the team for a product and engineering meetup. It was great to align on priorities, exchange ideas, and explore the next wave of products and opportunities together. Despite the busyness, I managed to squeeze in a couple of my NYC favourites — a visit to the Ghostbusters FDNY Ladder 8 firehouse and an early morning run around Central Park.

I ran the Brooklyn Experience Half Marathon in 1:57 and raised $3,906 for mental health in one week — beating my sub‑2 goal and surpassing fundraising target by 160%. A big thank you to the 40+ donors whose support made this possible, and gratitude to everyone who cheered me on. I finished in the top 19% for my age group and in the top 32% of all 27,000 runners, on a course with ~500 feet of elevation. Closed with a fast finish below average pace through the home stretch in Prospect Park — a great feeling after big hills. My 1:57 time is a 60% age‑graded performance, equivalent to running a 1:35 half marathon at peak age. The Brooklyn Experience Half Marathon has great vibes, beautiful neighborhoods (including several where I’ve lived), and a hilly, verdant final stretch. This was my eighth half marathon, building on four marathons and three half Ironmans. While running, I thought of family, friends, and colleagues who are fighting mental illness now, those we’ve lost, and those making progress. Active efforts, tragic losses, encouraging victories. I dreamed of a world without mental illness — no bipolar disorder, no schizophrenia, no PTSD, no depression. Perhaps research breakthroughs can accelerate cures in our lifetime. Conditions were demanding — 41 degrees, windy, 96% humidity — so I’m especially glad to have run sub‑two, kept an 8‑something pace, and brought home a new medal. Plus it was fun to run the race with a Kargo colleague; at the starting line we said: Ready. Set. Kargo! Thank you again for all the encouragement and generosity; it means so much. If you’d like to help push this farther, donations are still open, and every dollar supports the fight against mental illness: https://lnkd.in/eUUvVSHN

Kargo continues to grow our West Coast presence and we’re hiring across the LA office including a Senior Account Executive position. This is a group that shows up for each other and for our partners every day and that was recently recognized as Partner of the Year by our ThinkLA community. If you know someone who would be a great fit or if you are interested please check out Kargo Careers or reach out directly https://lnkd.in/gFJXB4q7

Kargo was part of the 10th year of ADGULLY ‘s DIGIXX last Friday in Gurugram. Day spent well with industry leaders and peers. While Cam Dinnie took the stage to share insights around AI in Advertising, I had the wonderful opportunity to moderate a panel discussion with some of the brightest minds in Marketing- Ashutosh Chaudharie , Abhishek Mehta Girish Kumar Girish Menon Sajit Gopal Utkarsh Mehta Thank you Bijoya A Ghosh , Sudhir Nair Sunil Pinto Vivek Mathur (VK) #adhya Sandeep Benny Abhishek T. Ishan Majmudar Mohnish Dodani Nicole Oei Bakul Varma

Great to kick off the Kargo Tech Blog! The inaugural post: our Machine Learning platform designed to support multiple teams deploying models, with deep dive into a key use case: dynamic product ads at high scale. These dynamic ads come from product catalogs that can include millions of product images, serving each user a relevant set based on key criteria. The ML platform powers this use case with strong performance. Here are five key takeaways from the paper authored by paddy raghav in our team: 1. Ultra-low latency is foundational. The architecture is built to deliver sub-20 millisecond predictions at high volume. Achieving this required end-to-end streamlining and an intense focus on latency minimization across the entire pipeline. 2. Real-time personalization required a rethink. Large lookup tables, offline processes, and batching approaches reached their limits as product catalogs scaled up. The system had to be redesigned from the ground up to support real-time learning and decisioning. 3. A unified ML platform enables scale. The new platform integrates model registry, feature store, inference layer, observability, and monitoring into a single system, tightly connected to business outcomes. This replaced fragmented scripts and ad-hoc pipelines. 4. Exploration-driven ranking changes the paradigm. The team reframed product selection as a contextual bandit problem, using adaptive decision-making to balance exploration and exploitation. This enables the system to iteratively identify the product images that drive the highest performance. 5. Multi-inference architecture enables future-proofing and optimizes cost The platform supports multiple inference modes — ONNX with C++ runtime for ultra-low-latency models, gRPC for general or batched workloads, and GPU-optimized serving for large or computationally intensive models like LLMs This is the first in a series of tech blogs from Kargo with more coming soon; glad to be sharing our learnings with the broader community and showcasing cool projects and products from the team. Special thanks to Andy Owens Mick O'Brien Simon Critchley Bradley Harkrader Alena Morris and the entire Machine Learning and Data Science team for helping to bring the blog post and core work to fruition. Stay tuned for a new post next week on Kargo's approach to re-inventing contextual targeting with AI authored by Andrew Underwood ! https://lnkd.in/eXiN_Ahw

Exciting addition to the Kargo Portfolio & Family! Welcome The Walt Disney Company streaming to our premium curated portfolio! Unmatched IP + beautiful creative + smart tech + Must CTV = powerful solutions to drive our partners & clients business! 🚀 📺 💥 !!