Software Engineer specializing in AI Systems, Machine Learning, and Backend Engineering.
Currently building production-grade Agentic AI, Voice AI, and Document Intelligence systems at Infosys, with a focus on scalable architectures, LLM orchestration, and cost-efficient AI infrastructure.
- ποΈ Production Voice AI for Banking & Insurance
- π€ Multi-Agent AI Systems
- π§ LLM Orchestration & Agentic Workflows
- π Intelligent Document Processing
- βοΈ AWS & Azure AI Services
- β‘ High-performance Python Backend Systems
Systems Engineer β Data Science & Machine Learning
Working on enterprise AI solutions including:
- Voice AI Kiosks
- Multi-Agent AI Architectures
- Document Intelligence Platforms
- LLM Optimization
- Retrieval-first AI Systems
- Production ML Pipelines
Python β’ Java β’ C++ β’ JavaScript
PyTorch β’ Scikit-learn β’ Pandas β’ NumPy β’ CNNs β’ Transfer Learning β’ Optuna β’ Machine Learning β’ Deep Learning
Amazon Bedrock β’ Azure OpenAI β’ Multi-Agent Systems β’ Pydantic β’ Open-source LLMs β’ Prompt Engineering
FastAPI β’ Spring Boot β’ Node.js β’ Express.js β’ REST APIs β’ asyncio
AWS β’ Amazon Bedrock β’ API Gateway β’ DynamoDB β’ Amazon S3
PostgreSQL β’ MySQL
React β’ HTML β’ CSS β’ Gradio β’ Streamlit
Enterprise multilingual Voice AI platform for Banking & Insurance featuring:
- Supervisor-based Multi-Agent Architecture
- Intent Routing
- Offline Speech Recognition
- Aadhaar & PAN Verification
- Biometric Authentication
- Privacy-first Design
- Cost-optimized LLM Orchestration
AI-powered document understanding system featuring:
- OCR Pipeline
- Metadata Extraction
- Retrieval-first Architecture
- Amazon Bedrock
- Azure OpenAI
- MCP-based Agentic Workflows
- Low-latency Structured AI Responses
Interactive deep learning playground comparing:
- ANN
- Optimized ANN
- CNN
- ResNet18
- Hyperparameter Tuned Models
Built using:
PyTorch β’ Gradio β’ Hugging Face Spaces
Machine Learning pipeline for wearable sensor data.
Highlights:
- 500+ Engineered Features
- 16 Experimental Configurations
- GridSearchCV
- 98% Test Accuracy
- Production LLM Systems
- Advanced Agentic AI
- Distributed AI Systems
- MLOps
- Model Optimization
- AI Infrastructure
- LinkedIn - https://www.linkedin.com/in/dikshantadlakha/
- GitHub - https://github.com/dikshantadlakha
- Email - adlakhadikshant9@gmail.com
Building reliable AI systems that move beyond demos into production.
