Skip to content
View umeshkedimi's full-sized avatar
:octocat:
GenAI + Backend & AI Platform Engineering
:octocat:
GenAI + Backend & AI Platform Engineering

Block or report umeshkedimi

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
umeshkedimi/README.md

Hi, I'm Umesh Kedimi 👋

AI Platform Engineer specializing in Agentic AISenior Software Engineer | Python | Backend & Distributed Systems


About Me

I'm a Senior Software Engineer with 9+ years of experience building production-grade backend systems and platforms.

My expertise lies in designing reliable APIs, distributed systems, and AI infrastructure that can be trusted in production. Today my primary focus is building Agentic AI platforms—systems that orchestrate LLMs, tools, workflows, and humans safely and reliably at scale.

I believe AI products succeed because of strong engineering, not just powerful models.


Current Focus

  • Agentic AI Platforms
  • AI Infrastructure Engineering
  • Multi-Agent Systems
  • Workflow Orchestration
  • LLM Application Architecture
  • Production AI Systems
  • Backend Platform Engineering

Tech Stack

Languages

  • Python
  • SQL

Backend

  • FastAPI
  • Pydantic
  • SQLModel
  • AsyncIO
  • REST APIs

AI & LLM

  • OpenAI
  • Anthropic
  • Google Gemini
  • MCP (Model Context Protocol)
  • RAG
  • Vector Databases
  • Embeddings
  • AI Tool Calling
  • Structured Outputs

Orchestration

  • Temporal
  • Background Workers
  • Event-Driven Systems
  • Async Processing

Data

  • PostgreSQL
  • Redis

Infrastructure

  • Docker
  • Kubernetes
  • Linux
  • GitHub Actions

Observability

  • Prometheus
  • Grafana
  • Logging
  • Metrics
  • Tracing

Engineering Principles

  • Build for production, not demonstrations.
  • Reliability is a feature.
  • Explicit architecture beats hidden complexity.
  • Simple systems scale better than clever ones.
  • AI should augment engineering—not replace engineering judgment.
  • Every system should be observable, testable, and maintainable.

What I'm Building

I'm currently focused on building production-ready AI infrastructure, including:

  • Agentic AI platforms
  • Multi-agent orchestration systems
  • AI gateways and platform services
  • LLM-powered backend applications
  • Workflow automation with Temporal
  • Reliable API platforms
  • Secure AI infrastructure

Career Direction

My long-term goal is to become a Staff-level AI Platform Engineer, designing the infrastructure that powers enterprise-scale Agentic AI applications.

I'm particularly interested in roles involving:

  • AI Platform Engineering
  • Agentic AI
  • Forward Deployed Engineering
  • AI Infrastructure
  • Distributed Systems
  • Backend Platform Engineering

Let's Connect

I'm always interested in discussing:

  • Python
  • Agentic AI
  • AI Platform Engineering
  • Distributed Systems
  • Backend Architecture
  • Production AI

Building the infrastructure that makes Agentic AI reliable, scalable, and production-ready.

Pinned Loading

  1. aicaicPublic

    Agentic Incident Commander — an autonomous reliability control loop that detects, investigates, and safely remediates production incidents with policy-gated, human-approved actions.

    Python 1

  2. prf-ai-pipelineprf-ai-pipelinePublic

    Production-grade Agentic AI platform for nonprofit fundraising campaigns (LangGraph multi-agent pipeline, MCP, RAG, human-in-the-loop)

    Python 2

  3. enterprise-agentic-ai-platformenterprise-agentic-ai-platformPublic

    Multi-tenant control plane for enterprise AI agents: onboarding a new assistant is a configuration change, not a deployment. Shared runtime for retrieval, orchestration, conversation, streaming, MC…

    Python 1

  4. ema_vwapema_vwapPublic

    Automated Nifty options bot: 5 EMA/VWAP crossover signals with trailing-SL risk management and a data-driven backtest log.

    Python 1 1

  5. askpaskpPublic

    ASKP — AI Secure Key Protocol. A secure platform to manage AI provider keys, issue scoped access tokens, and proxy requests without exposing real keys.

    Python 1