Browse categories
Explore
Fiverr Pro
English
$
USD
I will build a production ready ai agent with langgraph and tool calling
Akash L
About this gig
I build AI agents that survive contact with real users.
Most agent demos break the moment the model returns a malformed tool call. I build the layer that stops that: strict tool schemas, enum validation, cross-field correction, and graceful degradation instead of a stack trace.
What I have shipped: sole author of a multi-agent LangGraph assistant running in production at a US supply-chain platform. Natural language in, validated calls across tracking, geo and address APIs out, over a 125-parameter tool surface. It survived a full re-platform and still runs today.
WHAT YOU GET
A working agent in your repo, not a notebook. Tool schemas designed around your actual API. Structured output validation and fallback handling. FastAPI endpoint, Dockerfile and README. A call to walk you through the code.
STACK: Python, LangGraph, LangChain, OpenAI and open models, Pydantic, FastAPI, Qdrant, Docker, AWS.
Message me before ordering with what you are building. If it is not a fit, I will tell you.
Get to know Akash L
Akash L
AI Engineer, LLM Agents, RAG and MLOps
- FromIndia
- Member sinceMar 2020
Languages
Tamil, English
AI engineer with 4 years building LLM systems that run in production, not just demos.
I was sole author of a multi-agent LangGraph assistant for a US supply-chain platform. Plain-English questions turned into validated calls across a 125-parameter tool surface. It survived a re-platform and still runs today.
I build the unglamorous parts too: RAG pipelines, tool schemas, validation that degrades gracefully instead of crashing, and eval harnesses that catch regressions before your users do.
Python, LangGraph, LangChain, RAG, FastAPI, AWS.
Tell me what you are building.
FAQ
Which LLM providers do you work with?
OpenAI, Anthropic, and open-weight models run locally or via vLLM. I will recommend one based on your cost and latency needs rather than defaulting to the most expensive option.
Do I own the code?
Yes, fully. It is delivered into your repository with commit history, and you are free to modify, deploy and extend it however you like.
What if the agent gives wrong answers?
Expected at the start, which is exactly why evaluation matters. The Premium package includes a regression suite so you can measure accuracy and fix the specific failures instead of guessing.
