AI Engineer · GenAI & Agentic Systems · LLM Evaluation

Jibin Kunjumon

The model proposes.
My code decides.

Five production systems shipped across healthcare, fintech, and e-commerce, plus hands-on LLM evaluation and observability work on a live agentic RAG product — each built so the model can reason and draft, but the decision that matters runs through deterministic, auditable code.

5Systems Shipped
3Domains
250+Eval Cases Reviewed

B.Tech · CUSAT  ·  M.Acc · Deakin University, Australia

Jibin Kunjumon

Selected Work

Experience

AI Engineer (Contract) Jun 2026 – Aug 2026
Gistr · Remote

LLM evaluation & observability for a live agentic RAG product.

  • Built an isolated LLM evaluation system for answer accuracy, so testing could never affect live users.
  • Reviewed ~250 cases by hand and found the existing accuracy check was measuring the wrong part of the pipeline — redesigned it to catch answers the AI invented or couldn't support with a source.
  • Defined 12 agent tool-waste patterns and found two retrieval tools returning duplicate content, confirmed across 8 production sessions.
  • Created a hand-labelled golden dataset and version-controlled evaluation prompts to validate automated scores against ground truth.
LangSmith PostHog LLM-as-judge agent-as-judge LangGraph FastAPI
AI Engineer (Project-based) · Technical Lead Sep 2025 – Jun 2026
PM Accelerator · Kochi, India (Remote)

Backend and architecture lead across 3 concurrent production AI systems.

  • Led Python backend development across 3 concurrent production AI systems in healthcare, financial intelligence, and e-commerce.
  • Acted as Technical Lead across 3 cross-functional teams, owning architecture decisions and reviewing code through structured pull requests.
  • Worked with product managers, frontend engineers, and non-technical stakeholders to turn business requirements into deployed AI systems.
Python FastAPI LangGraph LangChain PostgreSQL GCP
AI Engineer Intern Sep 2024 – Aug 2025
GALTech Technologies · Kerala, India

Conversational AI systems on production FastAPI backends.

  • Built conversational AI systems on FastAPI backends with session management, conversation flow, and API integrations for production chatbot deployments.
Python FastAPI Conversational AI

More Work

Healthcare AI Deployed
Founding Engineer

RemiMinder

HIPAA-compliant audio-to-record pipeline on GCP. MedGemma extracts structured clinical fields from consultation audio; Gemini summarizes. ~60-second turnaround on a 15-minute recording.

Python FastAPI GCP MedGemma
Private
FinTech AI
Tech Lead

TickerPulse

6-stage financial ML pipeline: news ingestion → FinBERT sentiment → Z-score anomaly → GPT-4 alert summaries. JWT auth, watchlist API, React/TypeScript frontend.

Python FastAPI FinBERT GPT-4 PostgreSQL
GitHub
Agentic AI
Solo · Hackathon — 13 days

Autonomous Research Assistant

6-layer agentic system — Planner / Executor / Synthesizer agents via LangGraph. ChromaDB RAG + Tavily web search. Built bottom-to-top in 13 days solo.

LangGraph GPT-4o-mini ChromaDB Tavily Streamlit
GitHub
RAG / NLP
Academic

Discord RAG Bot

Production-quality Q&A bot with MongoDB Atlas vector storage, SentenceTransformers (384-dim), Azure OpenAI GPT-3.5 Turbo. 100% test accuracy, 69.2% avg token overlap.

Azure OpenAI MongoDB Atlas SentenceTransformers Discord.py
GitHub

Recommendations

What people I've worked with say

He developed our evaluation system from the ground up, creating an agent-as-judge pipeline that analyzes traces for tool usage, latency, cost, and failure reasoning, and incorporates human-reviewed golden datasets and in-depth investigations.
Arun Antony Founder @ Gistr.so Gistr · AI Evaluation
Read recommendation
Jibin has been the leader for entire LLM development for Mindgym and has demonstrated strong ownership and capability to develop backend from zero to 1 and has been awarded Tier 3: AI Rising Star Award!
Claire Hong Zhu Data Science Product Leader MindGym · LLM Development
Read recommendation
He was instrumental in architecting our 'Data Integrity Layer,' moving the project beyond simple LLM prompts into a strictly typed, contract-first system.
Sahal Alarabi Technical Product Strategist Revenue Recovery Engine · Backend & Architecture
Read recommendation
View all recommendations on LinkedIn

Certificates

PM Accelerator

Certified AI Engineer

Issued by Dr. Nancy Li · Certificate No. 72a972f9

Certificate on request
Anthropic

Claude & Agent Development

  • Claude with the Anthropic API
  • Claude Code in Action
  • Model Context Protocol (MCP)
  • Agent Skills

2026

Certificate on request
Educosys

Hands-on Generative AI

2025

Certificate on request

Let's build
something.

Open to full-time AI Engineer roles and interesting collaborations.

Ask about Jibin's work

Answers are drawn only from this portfolio's structured project and experience data.