I will test and evaluate your ai agent or llm app for reliability

S
sooby_senpai
S
sooby_senpai
Subodh Singh
test and evaluate your ai agent or llm app for reliability
Full Screen
test and evaluate your ai agent or llm app for reliability
Full Screen
Full Screen
test and evaluate your ai agent or llm app for reliabilitytest and evaluate your ai agent or llm app for reliabilitytest and evaluate your ai agent or llm app for reliability

About this gig

Is your AI agent calling the wrong tool, skipping approvals or giving confident wrong answers?


I'm the author of RegShield, an open-source regression-testing SDK for AI agents. I'll find where your agent or LLM app breaks and set up tests that catch it before your users do.


What you get:

  • Review of your prompts, tools and real failure cases
  • Scenario-based tests for tool order, arguments, approvals and forbidden actions
  • Accuracy and grounding checks with pass/fail scores
  • Regression tests in pytest that block bad changes in CI (GitHub Actions)
  • Optional LLM-judge checks for things rules can't catch

Works with LangGraph, LangChain, smolagents and raw OpenAI, Anthropic or Gemini SDK loops.


Message me before ordering with a short description of your agent and I'll recommend the right package.

  • Model expertise

    • Generative AI

  • Industry

    • Data analytics
    • Financial services

  • Programming language

    • Python
    • TypeScript

  • Language

    • English
    • Hindi

  • Technical expertise

    • Machine learning (Supervised, Unsupervised, Reinforcement)
    • Deep learning (Neural networks, GANs)
    • Natural language processing (NLP)
    • Computer Vision (Object detection, Image recognition)
    • Reinforcement learning (Decision-making systems)
    • Feature engineering and data processing
    • AI ethics and bias mitigation

Get to know Subodh Singh

Subodh Singh

AI Engineer building AI Agents, RAG Chatbots, LLM Evals and Web Apps

  • FromIndia
  • Member sinceApr 2025

  • Languages

    Hindi, English
AI engineer with 2+ years of full-time experience building production AI for defense (DRDO projects) and robotics. I build AI agents, RAG chatbots, multi-agent workflows and full-stack LLM apps with Python, LangGraph, LangChain, FastAPI and Next.js. Shipped: XGBoost threat-detection models and a 5-agent operator console (DRDO), RegShield (open-source SDK that tests AI agents in CI), Quillwick (multi-agent research assistant), DataLens (AI Q&A over Excel/CSV) and live client websites. Clean, tested code, daily updates, on-time delivery. Message me first for a clear plan and quote.

My Portfolio

Other AI Development Services I Offer

Related tags