| Role | Backend Engineer @ Takkada — building BharatVoice, a multilingual voice AI platform |
| Focus | RAG pipelines · agentic systems (LangGraph) · LLM fine-tuning · backend infra |
| Grinding | Competitive programming — Codeforces ~1700, 400+ DSA problems solved |
| Goal | Backend / Applied AI SDE at a top AI company |
- BharatVoice — multilingual voice AI platform, production RAG + LLM pipelines at Takkada
- Open Emergent — self-hostable AI app builder powered by a 9-agent parallel LangGraph mesh
- RAG Pipeline Optimizer — hybrid retrieval (BM25 + Cross-Encoder reranking) on Qdrant, with Redis caching + Langfuse observability
- AgentMesh — multi-agent orchestration framework built on CrewAI / LangGraph + the Anthropic API, speaking MCP
- CopyCraft — AI copywriting SaaS with 20+ conversion frameworks
Open to Backend / Applied AI roles — always happy to talk shop about RAG, agents, or CP.