Skip to content
View SkinnyFatBoy05's full-sized avatar

Block or report SkinnyFatBoy05

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
SkinnyFatBoy05/README.md

Kishore Srinivasan, AI Engineer and Software Engineer

Portfolio LinkedIn Email

I build AI and software systems that make operational work more reliable, inspectable and safe. My work sits where backend engineering, applied AI, data and product delivery meet.

I am completing a Master of Computer Science (Advanced Entry) at the University of Sydney in November 2026.

What I build

  • Guarded agentic systems with bounded tools, inspectable evidence and human review.
  • Backend services and data workflows using Python, FastAPI, SQL and PostgreSQL.
  • Applied ML systems with explicit evaluation, failure analysis and operational boundaries.

Selected engineering work

Production-style AI engineering platform for RAG, guarded read-only SQL, bounded LangGraph agents, evidence verification and human review.

Evidence: 114 backend tests, 9 frontend tests, 5 browser workflows and 40/40 deterministic evaluations of routing, tool safety and evidence integrity. This is not a live-model accuracy claim.

Python FastAPI LangGraph React PostgreSQL

Synthetic utility-data onboarding control plane for typed ingestion, reconciliation, data-quality gates and reviewable exceptions.

Python Airflow dbt Data Engineering

Synthetic fleet-operations MVP combining governed KPIs, data-quality controls, ML and evidence-linked AI recommendations.

Python FastAPI DuckDB ML Power BI

Point-in-time-safe weather and day-ahead power-market research with chronological evaluation and cost-aware backtesting.

Evidence: 51.6% lower final-holdout MAE than the declared seasonal-naive comparator, within the repository's stated research limitations.

Python Time Series ML Reproducible Research

Synthetic asset-finance workflow with explicit BPMN and DMN decisions, resilient workers, human review and auditable state.

Camunda 8 .NET C# React PostgreSQL

Safety-first telemetry validation and explainable operating-window recommendations for human decision support.

Python Data Validation Decision Support

Experience snapshot

  • Technical Co-Founder & Product Manager, KRSP Tech: stakeholder discovery, requirements, backend and product contribution, distributed delivery and release review.
  • Machine Learning Intern, Defence Research and Development Organisation: contributed to a commercialised Intelligent Character Recognition System through Transformer OCR work covering 22 Eighth Schedule Indian languages plus English, using 22 million text-image pairs overall across data preparation, training, inference and evaluation.
  • Master of Computer Science (Advanced Entry), University of Sydney: Artificial Intelligence, Data Science and Cybersecurity. Completing November 2026.

Technical toolkit

  • Languages: Python, TypeScript, JavaScript, SQL, C#, R
  • Backend and data: FastAPI, PostgreSQL, REST APIs, Pydantic, Docker, data pipelines
  • AI and ML: LangGraph, RAG, Transformer models, NLP, model evaluation, error analysis
  • Product and cloud: React, AWS, GitHub Actions, stakeholder discovery, release delivery

Research and recognition

Contact

If your team is building applied AI, backend platforms or automation systems, connect with me on LinkedIn or email kishoresrinivasan05@gmail.com.

Pinned Loading

  1. opsgraph-copilot opsgraph-copilot Public

    Production-style AI engineering reference platform for RAG, guarded text-to-SQL, LangGraph agents, evaluation and observability.

    Python

  2. gridbridge gridbridge Public

    Synthetic utility data onboarding control plane for meter, GIS, SCADA, DER and flex-market integration

    Python

  3. fleetguard-ai fleetguard-ai Public

    Audited synthetic fleet-operations MVP with DuckDB, FastAPI, ML, Power BI/DAX, Fabric/PySpark and governed AI.

    Python

  4. gridshock-research-lab gridshock-research-lab Public

    Point-in-time-safe weather and DE-LU day-ahead power-market research with walk-forward evaluation and cost-aware backtesting.

    HTML

  5. angleflow angleflow Public

    Camunda 8 BPMN asset-finance workflow built with .NET, C#, React, and Azure deployment examples.

    C#

  6. energy-operations-advisor energy-operations-advisor Public

    Safety-first Python telemetry validation and explainable flexible-load recommendations.

    Python