Available for work

Founder & BuilderVOICEAI.Full-Stack.Founder.

Co-Founder & CEO of Resonera.ai — building AI voice agents that handle calls for Indian SMBs & MSMEs in 25+ languages. Full-stack dev, ENTC undergrad.

Contact meMy work
Aditya Kumar Sah
~600ms
Voice Agent Latency
3+
Internships & Freelance work
2026
Founded Resonera.ai
Voice AI & Real-time Infra
Low-latency pipelines, WebRTC, ASR/TTS and conversational agents at scale.
Full-Stack Engineering
Next.js, Node.js, FastAPI — end-to-end products from API to pixel.
LLM & AI Systems
Prompt engineering, RAG, orchestration and self-hosted inference.
Founder & Product
From zero-to-one: incorporation, team building, GTM and brand strategy.
Voice AIFull-Stack DevLLM InfraWebRTCFounderUI/UX DesignNext.jsPythonResonera.aiOpen SourceReal-time AINode.jsFastAPIDockerProduct ThinkingVoice AIFull-Stack DevLLM InfraWebRTCFounderUI/UX DesignNext.jsPythonResonera.aiOpen SourceReal-time AINode.jsFastAPIDockerProduct Thinking
About me
ABOUT
Aditya Sah
Co-Founder & CEO @ Resonera.ai · Mumbai

Developer
Who Ships.

I'm a Electronics & Telecom Engineer student and the Co-Founder & CEO of Resonera.ai — we build AI voice agents that handle inbound and outbound calls for Indian SMBs and MSMEs in 25+ languages, far cheaper than a human telecaller. What makes ours different: they act, not just understand — taking real actions mid-call so the work actually gets done.

I write code that ships. We're building it all as one unified layer — from telephony to voice AI to campaigns and context — so businesses stop vendor hopping and run everything from a single platform. One platform, less headache.

When I'm not building, I'm exploring open-source tools, running local LLMs, or figuring out what the next version of human-computer interaction looks like.

Specialty
Voice AI Infra
Degree
Electronics & Telecom Engineer
Location
Mumbai, India
Focus
Full-stack + AI
Tech stack

What I Work
With.

From voice pipelines to production deployment — I own the full stack and the design layer.

Voice AI & Infra06
Conversational AILow-latency PipelinesWebRTCVADASR / TTSReal-time Streaming
Full-Stack Dev07
Next.jsReactNode.jsPythonFastAPIREST APIsWebSockets
AI / ML06
LLM OrchestrationPrompt EngineeringRAGData PipelinesOllamaOpenAI API
DevOps & Cloud06
DockerLinuxCI/CDGCPDNS & Email InfraSelf-hosted LLM
UI / UX Design05
System DesignComponent ArchitectureFigmaDesign SystemsResponsive Design
Founder Ops05
Startup OpsGST & ComplianceMarket ResearchTAM/SAM/SOMBrand Positioning
Portfolio

Things I've Built.

All repos
MVP Live
Resonera.ai

AI voice agents for Indian SMBs & MSMEs — handling inbound and outbound calls in 25+ languages, from appointment booking and sales to lead qualification and support. Agents that act, not just understand: they take real actions mid-call so the work gets done, at ~600ms latency and a fraction of a telecaller's cost.

PythonWebRTCASRTTSLLM
Visit resonera.ai
Live
NotchNux

A dynamic macOS-style Notch & dashboard for GNOME Shell — music, weather, webcam, calendar, notifications, and a file shelf, all accent-themable. Built for Linux on Wayland.

GNOME ShellJavaScriptLinuxWayland
MVP Live
timely-sync

Real-time productivity app for people who share their lives — co-founders, flatmates, couples. Shared calendar, to-dos, and end-to-end encrypted group chat with in-chat slash commands. Built with Expo (React Native) + Supabase.

React NativeExpoSupabaseTypeScript
MVP Live
lead-automation

AI-powered outreach toolkit (CLI) that finds and scores LinkedIn leads, drafts human-sounding messages, syncs to Google Sheets, and generates + posts X/Twitter content — grounded in a company knowledge base. Built for Resonera.ai.

PythonPlaywrightGroq LLMTavily
MVP Live
Call Center Pipeline

Data ingestion pipeline for call center audio — transcription, analytics, and training dataset generation at scale.

PythonWhisperMongoDBFastAPI
Personal
Local LLM Server

Self-hosted LLM inference running 2–3 models concurrently for low-cost prototyping.

OllamaPythonDockerGPU
MVP Live
PANNs Audio Analysis

Audio anomaly detection and classification using Pretrained Audio Neural Networks — classifies 527 sound categories, detects unusual sounds, and does frame-by-frame event detection, exposed via a FastAPI service and a Streamlit dashboard.

PythonPANNsFastAPIStreamlit
Career
WORK
Co-Founder & CEO
Resonera.ai2026 — present
  • Building AI voice agents for Indian SMBs & MSMEs — inbound & outbound calls in 25+ languages
  • Agents that act, not just understand: real actions mid-call (CRUD) at ~600ms latency
  • Unifying telephony, voice AI, campaigns & context into one platform — no more vendor hopping
  • Led company incorporation, GST registration, banking & compliance setup
  • Expanding the agent omnichannel across WhatsApp & SMS for small businesses
Contact
Let'sBuild.

Whether you want to talk voice AI, collaborate on a project, or just connect — my inbox is open.

GitHubXLinkedInresonera.ai
Personalsahaditya59@gmail.com↗
Workadityasah@resonera.ai↗