Piyush Gambhir.

Software Engineer II · Voice AI, Agentic Engineering & Infrastructure

I build production voice agents, and I build with agents. My workflow is agent-first: write the spec, delegate to coding agents, verify hard. It runs at roughly 4x my previous throughput across fifteen production repos. Stacks are interchangeable; I ship Go, Python, and TypeScript in the same week. What compounds is systems judgment, verification, and speed of adaptation.

Experience

Software Engineer II

Spyne · Full-time · Gurugram, India

May 2026 to Present
  • Rebuilt my development process around coding agents (spec-first delegation, parallel agent runs, and verification gates), quadrupling shipped output from ~60 to ~300 commits per month across 15 production repositories.
  • Scaling Spyne’s production conversational AI stack to support increasing traffic, concurrency, and reliability requirements.
  • Leading the migration of production voice agents from Vapi to LiveKit, giving us greater control over infrastructure, customization, and performance.
  • Reduced voice-agent infrastructure costs from $0.26 to $0.10 per minute, a reduction of approximately 62%.
  • Strengthening service resilience through disaster recovery, failure handling, and production readiness across the conversational AI platform.

Software Engineer I

Spyne · Full-time · India

Jan 2024 to May 2026
  • Built Spyne’s production conversational AI stack from zero to one on Vapi, powering dealership lead handling, follow-ups, and customer conversations, then began laying the groundwork for the transition to LiveKit.
  • Improved agent responsiveness and conversation quality through prompt optimization, streaming pipelines, turn-taking improvements, and infrastructure tuning.
  • Built the core telephony layer across Twilio, Vonage, and Telnyx, including caller identity, CNAM lookup, STIR/SHAKEN support, banned-number filtering, and A2P 10DLC workflows.
  • Shipped full-stack systems across inventory ingestion, dealer websites and CMS, CRM workflows, lead management, and automated customer follow-ups.

Projects

Document OCR

Local-first OCR toolkit for passports and Indian KYC documents, distributed as Python and TypeScript libraries with HTTP and Docker interfaces.

Telephony Voice Simulator

Provider-neutral test platform for voice agents, AMD, IVRs, PBX routing, live carrier behavior, recordings, and automated grading.

Reckon + Operations CLI Suite

Read-only production incident workspace plus Go CLIs for Grafana, Elasticsearch, Jenkins, Jira, CubeAPM, and Nginx Proxy Manager.

PDF Everything

Unified PDF console and REST API with nineteen tools for conversion, rendering, extraction, and document workflows.

Skills

Agentic Engineering

Claude CodeCodexSpec-first delegationMulti-agent orchestrationVerification loops

AI & Voice

LiveKit AgentsPipecatVapiLLMsRAG

Telephony

TwilioVonageTelnyxSIPWebRTC

Languages

TypeScriptPythonGoSQL

Backend & Data

Node.jsNestJSFastAPIPostgreSQLRedis

Infrastructure

AWSDockerCI/CDObservabilityGitHub Actions

Education

The NorthCap University

B.Tech (AIML), Computer Science

GPA: 8.55 / 10

2021 to 2025