Shubhankar Kahali

RSS Feed

I build things at the intersection of ML and people systems. I started in offensive security — reverse engineering binaries, writing exploit chains, finding zero-days across GoogleAppleMetaMicrosoftAmazonUber. That led to Hyperpage, a cybersecurity startup I co-founded — ML-driven threat detection replacing legacy signature-based approaches with neural nets for threat classification and automated incident response. Raised $12.8M, scaled to enterprise contracts, and exited through acquisition.


Now I train and deploy custom models — domain-specific transformers, graph neural networks, contrastive embedding systems — for problems where off-the-shelf models fail quietly. I also do niche headhunting, placing senior engineering and ML leaders at Fortune 500 companies. The intuition is the same in both: figuring out what will hold under real-world pressure and what will collapse the moment conditions shift.


My side project Glide Logo is where these converge. It's a career intelligence platform built on custom-trained models — a domain-adapted transformer continually pretrained on 4.2B tokens of career data, a heterogeneous graph neural network for relational matching across a knowledge graph of 82M+ companies and 230M+ jobs, custom contrastive embedding models with Matryoshka representation learning, and GRPO-aligned generative models for synthesis. The system uses speculative decoding with Eagle3 draft models for 2.4x latency reduction, INT8-quantized encoders via TensorRT, and drift-aware retraining pipelines that retrain on production feedback every 4-6 weeks. 444k+ users, 94% match accuracy, 81% interview success rate, $554M+ in total salary value delivered across 36.9k+ placements — with users averaging a 30% salary increase and offers closing in 16 days. Learn more about Glide →


Recent Posts

All Posts