Joined September 2016
- Tweets 2,132
- Following 2,438
- Followers 900
- Likes 16,551
512 Photos and videos
Jun 12
I was today years old when I noticed @posthog has a hedgehog companion just to prove that analytics doesn't have to be a slog.
Max is doing more for my motivation than any dashboard metric. π¦
2
4
5,454
Pratik Parmar retweeted
π We're excited to be a Day-0 launch partner for NVIDIA Nemotron 3 Ultra.
Deploy NVIDIA's latest open model for agentic AI on Simplismart
Our optimizations we deliver higher throughput than TensorRT MTP NVFP4.
Read more: lnkd.in/g7BrrNhy
#NVIDIA #Nemotron
1
5
9
224
Apr 22
Going live tomorrow to build a production-ready AI voice interviewer agent ποΈ
Sub-400ms latency, open-source models on @SimplismartHQ , voice agent orchestration with @LiveKit
5:30 PM PT π luma.com/jg93tpel
52
GoRevoke Interactive Sandbox is now live πβ¨
We make managing token approvals safer and effortless. Automate your token revocations with confidence.β¨β¨No more stress about exploits. Set your permissions once, and let GoRevoke keep your approvals in check.
5
4
8
425
Apr 6
Iβve been playing around with Qwen3-TTS for a real-time voice agent I'm building and it just blows my mind how insanely low the TTFB is π
When you're building conversational pipelines, latency is everything.
Here is what makes this TTS model so special 𧡠(1/n)
1
2
69
Apr 6
Instead of raw audio, Qwen uses a discrete multi-codebook tokenizer. It compresses sound into layers of discrete codes representing semantics and acoustics.
To clone a voice, it just needs a 3s reference audio to create an acoustic prompt blueprint. (2/n)
1
45
Apr 6
Real-Time Generation β‘
To speak, the autoregressive language model uses that blueprint to predict new audio codes step-by-step based on your text.
A decoder then instantly translates those discrete numbers back into a listenable, continuous waveform. (3/n)
17
Pratik Parmar retweeted
Mar 31
76
418
2,867
622,915
Feb 12
Starting in just a few hours π₯
Excited to walk through how weβre building an end-to-end invoice automation pipeline using VLMs and n8n.
Join us for InferSights by Simplismart.ai, to learn how to automate invoice processing end-to-end using VLMs and n8n workflows. Build a real-world pipeline that extracts, validates, and structures invoice data at scale.
π
12 Feb | 6:00 - 7:00 PM IST
72
Feb 6
RT @ShivrattanDhil1: Deeply disturbing to see a biker lose his life after falling into a deep, open pit at night in Janakpuri, Delhi.
Hisβ¦
1,633
24 Dec 2025
DocsForAI.dev made it to my @ChatGPTapp wrapped!
GPT babaβs advice for 2026: build in public β¨
4
142
22 Dec 2025
Bengaluru is so advanced that even autos come with @MCP_Community support π€π¨βπ»
@peakbengaluru
3
40
18 Dec 2025
You know you're surrounded by Al nerds when laptop stickers casually feature the entire Transformer architecture π€
3
48
12 Dec 2025
Ready to roll π€©
9 Dec 2025
Join us for InferSights by Simplismart and @pipecat_ai π
Learn how to build real-time voice agents with Pipecat and optimize open weight models for fast, production-ready voice applications.
π
Dec 11 | β° 5 PM PT
Register here: luma.com/cvnyuvrq
1
8
130
Pratik Parmar retweeted
2 Dec 2025
Want to build a voice agent with <400 ms latency? π€
Join InferSights #2 on Dec 11 to learn how we deploy and optimize real-time voice agents with Simplismart.
π
11 December 2025
RSVP: luma.com/cvnyuvrq
#GenAI #webinar
1
2
110