Databases, Apis, PostgreSQL, Node.js, React, Typescript, Python, AWS, RAG embeddings, AI workflows, LLMs, Agents, structured outputs, serverless functions, AI evaluation pipelines, Next.js, queues, workflow systems

data engineering , Python, Deep Learning, Nlp, Tensorflow, Pytorch

Tensorflow, Pytorch, Python, semantic search, Generative AI LLMs, ONNX, Model Quantization, Prompt Engineering, ChromaDB, Optimization, Edge On-device AI Deployment, Vector Databases, RAG Pipelines, GGUF, Embedding Models, Transformer Architectures, LLM Fine-Tuning, Pinecone, FAISS, llama.cpp, TensorRT

Tensorflow, Pytorch, Python, Optimization Techniques, ONNX, ChromaDB, Embedding models, Vector databases, INT8, Transformer architectures, model quantization, LLM fine-tuning, GGUF, 4-bit, semantic search, Pinecone, RAG architectures, FAISS, prompt engineering, llama.cpp, TensorRT

Apis, React, Distributed Systems, Python, CrewAI, LangChain, LLMs, RAG vector databases, CoT, Tree-of-Thoughts, tool-augmented LLM workflows, agent frameworks, reasoning patterns, Semantic Kernel, AutoGen, OpenAI SDK, async workflows, Hugging Face transformers, hybrid retrieval approaches

Prometheus, Grafana, Cuda, Azure ML, Pytorch, Docker, Kubernetes, Python, LangChain, DeepSpeed, Hugging Face Transformers, MLflow, Llm, SageMaker, GCP AI Platform, FAISS

react.js , Machine Learning, Data Modeling, ELT, TailwindCSS, Python, Etl, LangChain, Generative AI, Data Pipelines, RAG frameworks, HuggingFace, Microservices Architecture, OpenAI, Prompt Engineering

Azure, Tensorflow, AWS, Distributed Systems, Pytorch, Python, Kubernetes, Docker, Gcp, fine-tuning LLMs, model optimization, deployment strategies, vector databases, embeddings, Weaviate, Hugging Face Transformers, Pinecone, FAISS, prompt engineering

Pytorch, Python, Multi-GPU systems, Hugging Face, ESPnet, Mixed precision, LoRA adapters, CTC RNN-T, Speech LLM model training, Distributed training, ASR fine-tuning, nemo, Evaluation frameworks, MLflow logging

Python, negotiation workflows, Data Analysis, transcript analysis, API work, failure pattern identification, LLM integration, sentiment-based workflow tuning, per-call token cost optimisation, prompt design