Why hire me

I ship voice agents to production Lead the AI & real-time stack of two AI interviewers at InCruiter: 1,000+ concurrent interviews in 42+ languages on LiveKit Agents, Vapi telephony and Azure OpenAI, with feature-flagged releases to LiveKit Cloud.

I fix the frameworks I build on 16 merged PRs in other people's repos, 7 of them in LiveKit (agents, python-sdks). Every one starts with a test that fails on main: proof first, smallest fix second.

I understand the models from the paper up 7 papers rebuilt from scratch in PyTorch, each with a check script: Transformer, GPT + BPE, Whisper, Conformer + RNN-T, LLM.int8/GPTQ/FlashAttention, LoRA/DPO/GRPO and speculative decoding.

Looking for: Voice AI Engineer · Agents / Applied AI Engineer · Real-time ML · Forward Deployed Engineer
Based in: India (IST) · remote-first, comfortable overlapping with US and EU teams
📫 hassankazinasik@gmail.com · resume on request

Proof of work: merged open source

        

Project	Merged fix	Why it mattered
LiveKit Agents	#7741 · #7742	the call recorder no longer hangs when its encoder fails · the TTS markup tokenizer survives stray closing tags
LiveKit python-sdks	#860 · #861 · #862 · #864 · #865	webhook sha256 check · Twirp base path behind proxies · AudioMixer survives a slow input · correct I420A/I422/NV12 plane layout · canSubscribeMetrics grant
Piper TTS (Open Home Foundation)	#328	Hebrew voices get the Hebrew phonemizer in PyTorch inference
Mastra	#27102	Sarvam listen() uses the configured model; three more of my bug reports fixed by the team
RealtimeSTT / RealtimeTTS	#329 · #330 · #332 · #383	Hindi/Bengali transcripts keep their final full stop · feed_audio on tiny chunks · client auto-detects language · ElevenLabs voice settings sent
stream2sentence	#16 · #17 · #18	Urdu sentences split at ۔ · the default tokenizer splits Hindi, Chinese and Arabic at । 。 ؟ · Chinese/Japanese sentences joined without a stray space

All merged PRs · In review now (LiveKit, Pipecat, NVIDIA NeMo, sherpa-onnx, Piper)

About me I build real-time voice agents, and I keep asking one question: why do voice agents break on real Indian speech, and how do we fix it? 🏢 Now: Product Developer at InCruiter, leading the AI & real-time stack behind InCBot and IncScreen 🌍 Before: AI Engineer at myRide (Boston, fully remote) 🐛 Found in prod: English-only guards scored Devanagari 0.39 vs a 0.5 threshold; Marathi STT hallucinated in 10 of 15 turns 🔭 Building: an English/Hindi/Marathi voice agent on open models, CPU only 🏆 9× hackathon winner · CodeChef 3★ · LeetCode Knight

How a production voice agent works

📄 Papers I rebuilt from scratch

	Paper	Repo	What it shows
🧠	Transformer (Vaswani 2017)	attention-is-all-you-need-hindi	English → Hindi on 1.36M sentence pairs
🧠	GPT + BPE	gpt-from-scratch	12M-param GPT with its own tokenizer
👂	Whisper (Radford 2022)	whisper-from-scratch	exact audio front-end, WER/CER on English and Hindi
👂	Conformer + RNN-T	conformer-rnnt	streaming ASR with transducer loss and beam search
⚡	LLM.int8 · GPTQ · FlashAttention	quant-flash	measured perplexity on Qwen2.5-0.5B, Triton kernel
🎯	LoRA · DPO · GRPO	post-training	preference training that makes a model answer Hindi in Hindi
🚀	Speculative decoding	speculative-decoding	Leviathan et al. 2023, Algorithm 1 on a hand-written Llama

Toolbelt

🎙️ Voice AI

🧠 ML & LLMs

⚙️ Engineering

🕒 Recently pushed portfolio — My portfolio (hassankazi.vercel.app) and the LiveKit voice agent behind its… · 2026-10-28 The-Consistent-Coders — This is a inituative for beginners , freshers and freelancers to crack their… · 2026-10-25 bhai — Give it a repo, a ticket, or a task. Reproduces first, smallest fix, proves it,… · 2026-10-17

🖥️ hassan@voice-lab

GitHub activity

Building voice AI for your users? Let's talk.