https://therohityadav.com Rohit Yadav is a data & AI engineer writing about data platforms, LLM-powered automation and agentic systems. en https://therohityadav.com/blogs/the-two-gaps-behind-almost-every-bad-ai-output https://therohityadav.com/blogs/the-two-gaps-behind-almost-every-bad-ai-output Sun, 21 Jun 2026 10:53:35 GMT An LLM rarely disappoints because it cannot think. It disappoints because of what we forgot to place in front of it. The model is only ever as good as the world we hand it, and when the output arrives malformed or hollow or beside the point, the failure is almost always upstream of the machine […] https://therohityadav.com/blogs/the-ai-bubble-and-the-railroad https://therohityadav.com/blogs/the-ai-bubble-and-the-railroad Thu, 11 Jun 2026 09:14:12 GMT Every few decades, the world convinces itself that a new technology will change everything. And the curious part is that the world is usually right. The railways changed everything. The internet changed everything. The aeroplane shrank the planet. Yet the investors who financed these revolutions were, more often than not, left holding the bruises. Technology […] https://therohityadav.com/blogs/the-case-against-giving-your-ai-agents-job-titles https://therohityadav.com/blogs/the-case-against-giving-your-ai-agents-job-titles Thu, 04 Jun 2026 18:15:00 GMT When an agent fails, we blame its intelligence. The real constraint is quieter: not whether it can reason, but whether it has the room to. https://therohityadav.com/blogs/building-support-agent-without-a-vector-database https://therohityadav.com/blogs/building-support-agent-without-a-vector-database Thu, 04 Jun 2026 08:30:59 GMT I built Support Agent on a thousand support tickets without a vector database. At this scale, brute-force cosine similarity in numpy is faster, simpler, and exact. Here is why the absence is a feature. https://therohityadav.com/blogs/when-one-write-becomes-two-the-transactional-outbox-pattern https://therohityadav.com/blogs/when-one-write-becomes-two-the-transactional-outbox-pattern Sat, 30 May 2026 07:56:34 GMT Place an order, then publish the event. Two lines of code, one silent data-corruption bug. How the transactional outbox pattern closes the gap between what your database commits and what the rest of the system hears about, and why the fix is just a row. https://therohityadav.com/blogs/sqlalchemy-and-alembic https://therohityadav.com/blogs/sqlalchemy-and-alembic Fri, 22 May 2026 06:43:30 GMT You launch with three columns. Six months later, the schema has been altered a dozen times by three different engineers across four environments, and nobody can describe with confidence what the production database actually looks like. This is the problem Alembic was built to solve, and the one SQLAlchemy, despite being the more famous of the two, cannot touch. https://therohityadav.com/blogs/monitoring-a-dlt-pipeline-with-datadog-on-databricks-free-edition https://therohityadav.com/blogs/monitoring-a-dlt-pipeline-with-datadog-on-databricks-free-edition Tue, 05 May 2026 22:28:24 GMT Why the built-in Datadog integration falls short on serverless, and how custom DLT hooks plus Terraform-managed monitor fill the gap. https://therohityadav.com/blogs/what-i-learned-building-a-smart-grid-pipeline-on-databricks-free-edition https://therohityadav.com/blogs/what-i-learned-building-a-smart-grid-pipeline-on-databricks-free-edition Sat, 02 May 2026 20:26:51 GMT I built a smart grid meter pipeline on Databricks Free Edition with medallion architecture, DQX validation, SCD modeling, and Terraform IaC. Here is what worked, what broke, and what I learned. https://therohityadav.com/blogs/why-you-need-an-ai-harness https://therohityadav.com/blogs/why-you-need-an-ai-harness Sun, 12 Apr 2026 00:54:40 GMT Think the LLM is the final product? Think again. Raw models are powerful but unpredictable. Like a horse without a harness. Learn why a well-engineered AI harness is the secret to turning non-deterministic models into reliable, enterprise-ready tools.