Anish J.

I help companies build full-stack AI systems — from powerful LLM workflows to scalable Python APIs and cloud-ready infrastructure. I bring together the worlds of AI/ML and software engineering, delivering production-grade systems that are fast, reliable, and extensible. My work spans everything from LLM fine-tuning to Python backend APIs with FastAPI, LangChain, Redis, and AWS. Services I Offer: 1. LLM Development & RAG Systems • Build intelligent AI agents and assistants using Mistral, Llama, GPT-2, OpenAI, Groq APIs • RAG pipelines with LangChain, Pinecone, Neo4j, LlamaIndex • LangGraph orchestration, ReAct & Tree of Thought prompting • Structured output support (JSON/XML), tool calling, streaming, function calling 2. Python Backend & API Engineering • FastAPI-based scalable APIs with: • Async endpoints & modular routing • JWT authentication, rate limiting • Redis caching (prompt optimization with SHA256) • Pydantic schema validation with type hints • Dockerized microservices for ML models and tools • GitHub Actions for CI/CD automation • Integration with frontend clients or mobile apps 3. LLM Fine-Tuning & Evaluation • Train open-source models using: • PEFT (LoRA, QLoRA) • Instruction tuning & RLHF • Model monitoring & evaluation (BLEU, ROUGE-L, F1, MRR) • Bias & toxicity filtering pipelines with MLflow tracking 4. AI Infrastructure & Cloud Deployment • Host models using vLLM and HuggingFace Transformers • Use Apache Spark for ingestion and streaming pipelines • AWS expertise: SageMaker, Lambda, ECS, S3, CloudWatch • Build real-time indexing and document processing systems My Tech Stack Includes: • Languages: Python, Bash, C/C++ • Frameworks: FastAPI, Transformers, LangChain, LlamaIndex • Libraries: Pydantic, Redis, NumPy, HuggingFace, Pinecone • Tools & Infra: GitHub Actions, Docker, AWS, MLflow, Apache Spark • MLOps: Model lifecycle, inference optimization, structured deployment