Abhishek Rajput
Backend & Systems Engineer
Robotics & AI Infrastructure
Building from R&D to Production
./hello.sh
Hey, I'm Abhishek 👋
Backend & Systems Engineer with 1.5+ years building AI platforms, autonomous robotics systems, and industrial software. I solve problems around distributed systems, real-time infrastructure, and taking products from R&D to production.
Recently Shipped:
- Production RAG Platform (intelligent routing, 30% cost reduction)
- Robot Control System with 3× lower latency
- Multi-Tenant IIoT Platform processing 60,000+ sensor tags
./experience-log
- Architected a production-grade RAG platform enabling conversational interaction with user-uploaded documents through semantic retrieval and source-aware responses.
- Built an end-to-end retrieval pipeline spanning OCR extraction, semantic chunking, TEI embeddings, Qdrant vector search, cross-encoder reranking, and Llama 3 generation.
- Implemented distributed ingestion and retrieval workflows using Django, Celery, Redis, and WebSockets for asynchronous processing and real-time updates.
- Designed intelligent query orchestration through intent routing, contextualization, and query expansion, achieving ~1.5–2s end-to-end response latency.
- Django
- Qdrant
- Llama 3
- Celery
- Redis
- Next.js
- TypeScript
- Supabase
- TEI
Nexus Automech
- Python
- Django REST Framework
- ROS2
- WebSockets
- Modbus TCP
- YOLOv8
- Redis
- Celery
- PostgreSQL
Rapidops
- React.js
- Next.js
- JavaScript
- html
- CSS
- Tailwind CSS
BISAG-N
- Python
- Deep Learning
- Computer Vision
- OpenCV
- TensorFlow / PyTorch
./projects (3)
Production-grade document intelligence platform for conversational search, study workflows, and knowledge retrieval. Built to solve a common problem in document-chat systems: retrieval quality. Rather than sending every retrieved chunk to the LLM, the system uses query orchestration, reranking, and relevance thresholds to ensure responses remain grounded in source material.
Key Results
- Low-Latency Generation: Achieved ~1.8s average end-to-end latency.
- Token Optimization: Reduced unnecessary LLM calls by 60–70% through intent-aware routing.
- Live Synchronization: Implemented real-time document processing updates via WebSockets.
- Learning Engine: Built support for AI-generated flashcards and study questions.
Engineering Highlights
- Async Ingestion Pipeline: Structured full background extraction tasks running OCR → semantic chunking → embeddings → indexing.
- Query Orchestration: Built a comprehensive execution route handling contextualization → intent routing → expansion → retrieval → reranking → generation.
- Local Embedding Inference: Deployed local embedding models via TEI delivering ~5ms execution times instead of external API bottlenecks (~150ms).
- Strict Filtering: Integrated cross-encoder reranking layers to ensure low-quality or irrelevant chunks never reach generation phases.
- Distributed Architecture: Tied systems together cleanly using Django, Celery, Redis, Qdrant, and WebSockets.
- Django
- DRF
- Celery
- Redis
- PostgreSQL
- Qdrant
- TEI
- Cross-Encoder
- Groq
- Llama 3
- Django Channels
- Next.js
- ROS2
- Django
- Redis
- WebSockets
- Modbus TCP
- YOLOv8
- OpenCV
- UWB
- 3D LiDAR
- Django
- PostgreSQL
- Celery
- Redis
- WebSockets
- Modbus TCP
- RBAC
- Multi-Tenancy
./publications (3)
systems.hashnode.dev
Building a Production-Grade RAG System: Architecture, Trade-offs, and Lessons Learned
- Published on
Modernizing Defense Robotics: From PLC-Heavy Control to a Sub-150ms Web-Orchestrated HMI
- Published on
IIoT Telemetry Core Ingestion Engineering: Processing 60K High-Frequency Tag Packages at Scale
- Published on