I'm Saksham Pathak, an AI Engineer and M.Sc. AI & ML student at IIIT Lucknow, focused on building intelligent systems that can listen, reason, retrieve knowledge, and take action. My work spans Voice AI, LLM Agents, RAG Pipelines, Fine-Tuning, and AI Automation, with a strong interest in creating real-time AI experiences that feel natural and human. Currently, I'm exploring:
I enjoy turning ambitious AI ideas into practical systems that people can interact with, trust, and use every day. |
|
| Domain | Stack |
|---|---|
| Voice AI | Whisper · Deepgram · WebRTC · VAD · STT / TTS · Twilio |
| LLM & Agents | LangChain · LangGraph · LlamaIndex · OpenAI · HuggingFace |
| Training | PyTorch · LoRA · QLoRA · PEFT · Transformers |
| RAG & Vector | FAISS · ChromaDB · Pinecone · Semantic Retrieval |
| Infra | FastAPI · Docker · GCP Cloud Run · Vercel |
01 — Real-Time 3D Conversational Avatar
Speech → LLM → TTS with sub-second latency and neural lip-sync.
PyTorchWebRTCTTS3D
02 — DocuMind AI · Document Intelligence
Q&A across massive document collections. Semantic chunking + vector-backed retrieval.
LangChainFAISSChromaDBFastAPI ·
03 — FALCON · AI Fake News Detector
Multi-stage fact-checking pipeline: classify → verify → explain.
LangChainOpenAIFastAPI ·
04 — parth-dl · Open Source CLI
Zero-dependency Instagram media downloader. No login, no API key. Published on PyPI.
PythonCLI ·
05 — 100 Best GenAI Projects 2025
Curated 100+ high-impact open-source LLM, RAG & Agent projects.
LLMRAGAgents ·
- GitHub Docs (Official GitHub Repository)
- Merged PR #45118: Fix duplicated word in image alt text for Copilot access docs
- 🔗 github/docs#45118
"The goal isn't to build models. The goal is to build intelligence that matters."