Conversational AI
Prompt systems, dialogue design, guardrails and evaluation for agents that behave consistently.
AI Engineer · Voice Systems · Consulting
I build AI systems that can listen, reason & respond.
Software engineer focused on conversational AI, ML systems and production-grade voice experiences — from real-time audio pipelines to agents that hold useful conversations.
Available for select AI consulting projects
01 / What I do01 02 03
Engineering intelligence into real products.
From model behavior to infrastructure, I work across the full stack required to ship reliable AI experiences.
Voice & Realtime
Telephony, streaming audio, STT/TTS, endpointing and latency-sensitive orchestration.
ML Engineering
Production inference, data flows, observability and pragmatic model integration.
02 / Selected work01 02 03
Systems, not demos.
Voice AI platform
Production-grade agent infrastructure
Outbound and inbound calling, real-time streams, scheduling, analytics and provider integrations.
Conversation quality
Evaluation & behavior engineering
Prompts, structured checks and conversation analysis designed around measurable outcomes.
Open source
Latency tools & agent evals
Public experiments in realtime tooling and honest agent measurement — including this site.
03 / Writing
Notes from the workbench.
Practical essays on voice AI, agent behavior and building production systems.
All posts →04 / About
Engineer first. Systems thinker always.
I'm Zeeshan — a software engineer based in Srinagar, working with teams everywhere. I spend my days building voice-AI platforms: realtime audio pipelines, agent orchestration, and the evaluation infrastructure that keeps conversational systems honest in production.
I care about the twenty milliseconds nobody notices and the guardrails nobody sees. When something breaks in an interesting way, it usually ends up on the blog.