Writings on Software Engineering, Software Architecture, Generative AI and LLMs
I was reviewing our voice calls today and discovered that in one of voice agents when user was saying “Now” our system was handling it as “No”. After we reset their password we ask them “Would you like to try logging in with this temporary password now, or would you prefer to do it later?”. … Continue reading "STT Disambiguation in Voice Agents: It is “Now” Not “No”"
I spent the second half of 2025 building a voice agent platform for a company that provides IT support services to higher education institutions. The platform is now live and handling calls from students and staff across multiple universities. We built a multi-tenant system where students call a phone number, speak with an AI agent, … Continue reading "Building Production-Ready Voice Agents"
I have a Python HTTP client built using the httpx library. The client worked fine when I was working with the test environment, but as soon as I pointed it to the prod environment, the request started failing with an HTTP 403 error. To be 100% sure, I tested the same request using the Bruno … Continue reading "Debugging failed HTTP request with Claude Code"
I was reading a paper by the Google DeepMind team on how they trained Gemini Embedding, a state-of-the-art, unified embedding model. This is the second paper I’ve read this month on training embedding models. Last week, I read about how the Jina embedding model was trained. The Jina embedding paper was thin and lacked details, … Continue reading "Notes from Gemini Embedding Paper"
I was working on a problem where I needed to extract information from hotel tariff sheet PDF documents. These documents provide details on seasonal room rates, occupancy terms, and related supplements. They serve as standard reference material for travel agents, tour operators, and partners when contracting accommodations. Below is a screenshot of a synthetic document … Continue reading "Comparing…
I was going over the code base of mini-swe-agent today. The core agent loop is 100 lines long. All agentic framework does something similar. Interesting facts about mini-swe-agent: The Mini-SWE-Agent operates in a continuous loop, iteratively solving problems by querying an LLM for actions, executing bash commands, and observing results until the task is complete. … Continue reading "Notes on…
I was reading state of AI in Business 2025 report today and below resonated with me. I am also building enterprise Generative AI products and found it useful. Memory is becoming critical for succeeding at Generative AI products. People expect Generative AI products to become smart. I have listened to multiple talks this year. These … Continue reading "How executives select GenAI vendors"
Today I was going over a paper by Microsoft Research team on how AI is impacting professsional work. This paper was published in July 2025. They analyzed 200k anonymized and privacy-scrubbed conversations between users and Microsoft Bing Copilot to understand how generative AI impacts different occupations and work activities. They seperated analysis into two distinct … Continue reading "Paper:…
Google recently released Gemma 3 270M, a remarkably compact 270 million parameter language model that promises efficient AI capabilities in a tiny package. As someone building AI voice agents, I was immediately interested in testing whether this model could handle one of my simplest but frequent use cases: generating message variations for conversational AI. For … Continue reading "I Tested Gemma…
Today, I was browsing Hacker News when I stumbled upon an interesting project: coderunner-ui. The premise was compelling – a local-first AI workspace that lets you chat with LLMs and execute generated code in isolated environments, all without sending your data to the cloud. As someone who’s always looking for tools that respect privacy while … Continue reading "Making coderunner-ui work with…