I will build rag pipelines and ai powered backend systems

M
mathan_teqbahn
M
mathan_teqbahn
Mathan K
build rag pipelines and ai powered backend systems
Full Screen
build rag pipelines and ai powered backend systems
Full Screen
Full Screen
build rag pipelines and ai powered backend systemsbuild rag pipelines and ai powered backend systemsbuild rag pipelines and ai powered backend systems

About this gig

AI Backend Engineer | RAG Pipelines · LLM Integration · Agentic Systems


I'm a Senior Software Engineer with 12+ years building production AI systems. I'm the sole architect of Teqit a live enterprise platform with RAG pipelines, on-premise LLM inference, and agentic microservices serving 10+ clients.


What I build for you:

- RAG pipelines document ingestion, vector search, LLM answer generation

- LLM API integration OpenAI, Claude, Gemini, or any hosted model

- Local LLM inference with Ollama / llama.cpp (cut AI costs ~70%)

- Agentic workflows tool calling, multi-step reasoning, decision pipelines

- FastAPI or Scala backends via REST or gRPC

- Hybrid routing smart switching between local and cloud models


Why me?

I've done this in real production not tutorials. I've cut AI costs 70% with local inference, built generative video avatar pipelines, and designed agents that independently reason over enterprise knowledge bases.


Tech: Python · FastAPI · Ollama · OpenAI API · Akka/Scala · Docker · AWS


Message me before ordering I'll review your use case first.

  • AI engine

    • ChatGPT
    • Claude
    • Hugging Face
    • Ollama

  • Tools & frameworks

    • LangChain
    • LlamaIndex

  • AI solution type

    • Rag system

  • Purpose

    • Customer support
    • Document processing
    • Customer service

Get to know Mathan K

Mathan K

Senior AI and Backend Engineer

  • FromIndia
  • Member sinceSep 2026
  • Avg. response time1 hour

  • Languages

    English
Senior Software Engineer with 12+ years building production AI systems, distributed backends, and cross-platform mobile apps. I'm the sole architect of Teqit — a live multi-tenant enterprise platform with RAG pipelines, on-premise LLM inference (Ollama), and agentic microservices serving 10+ enterprise clients. I build RAG pipelines, LLM integrations, agentic workflows, and FastAPI/Scala backends — cutting AI costs ~70% with local inference. I review every project personally before starting. Message me first.

Related tags