https://abhishekbagalkot.com/Recent content on Abhishek BagalkotHugoen-usFri, 18 Sep 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/hello-world/Fri, 18 Sep 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/hello-world/The first post — a check that the whole pipeline works: Markdown in, Hugo build, GitHub Actions deploy, live site out.https://abhishekbagalkot.com/posts/why-route-optimisation-projects-succeed/Sat, 12 Sep 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/why-route-optimisation-projects-succeed/The ones that work don&rsquo;t have better solvers. They treat the first quarter as constraint discovery, and they scope against a number someone is already accountable for.https://abhishekbagalkot.com/posts/what-a-camera-cannot-see/Mon, 24 Aug 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/what-a-camera-cannot-see/Before any computer vision project, write down what the camera physically cannot observe — and check whether the thing you care about is on that list. It often is.https://abhishekbagalkot.com/posts/the-last-mile-is-a-people-problem/Wed, 05 Aug 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/the-last-mile-is-a-people-problem/Most last-mile inefficiency isn&rsquo;t in the routing. It&rsquo;s in the twenty minutes before the van leaves and the things drivers know that the system doesn&rsquo;t.https://abhishekbagalkot.com/posts/pilot-purgatory/Sat, 18 Jul 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/pilot-purgatory/Enterprises don&rsquo;t struggle to start AI pilots. They struggle to end them. The pilot that neither fails nor scales is the default outcome, and it&rsquo;s a structural problem, not a technical one.https://abhishekbagalkot.com/posts/what-pl-owners-want/Thu, 02 Jul 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/what-pl-owners-want/They are not buying a percentage. They are buying the ability to explain a number to someone above them — which is a different product than the one most vendors pitch.https://abhishekbagalkot.com/posts/counting-is-not-understanding/Sun, 14 Jun 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/counting-is-not-understanding/A vision system that counts correctly and explains nothing is a more expensive clipboard. The interesting output isn&rsquo;t the count — it&rsquo;s the variance nobody could see before.https://abhishekbagalkot.com/posts/notes-on-teaching-executives/Thu, 28 May 2026 00:00:00 +0000https://abhishekbagalkot.com/posts/notes-on-teaching-executives/Senior people don&rsquo;t need the concept explained. They need permission to admit which part they never understood, and a worked example from their own P&amp;L.https://abhishekbagalkot.com/about/Mon, 01 Jan 0001 00:00:00 +0000https://abhishekbagalkot.com/about/<img class="portrait" src="https://abhishekbagalkot.com/img/abhishek.jpg" alt="Abhishek Bagalkot" width="200" height="200"> <p>I work at the intersection of technology, operations and business — mostly on the unglamorous middle, where a working model meets a working day.</p> <p>Most of what I do falls into two buckets. The first is computer vision for enterprise clients: cameras pointed at industrial processes, and the much harder question of what anyone should do with what they see. The second is route and transport optimisation — I run <a href="https://runtyson.com">Tyson</a>, a logistics product, and write a series on dynamic route optimisation aimed at the people who carry the P&amp;L rather than the people who build the models.</p>