Skip to content
View sadiq-git's full-sized avatar
🎯
Focusing
🎯
Focusing
2 followers · 0 following

Achievements

Achievement: Arctic Code Vault Contributor

Achievements

Achievement: Arctic Code Vault Contributor

Organizations

@M7r0bot

Block or report sadiq-git

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
sadiq-git/README.md
sad7q

site linkedin role loc

$ whoami
sadiq // forward deployed, agentic ai

$ cat /etc/role
Agentic AI Engineer @ Tata Consultancy Services
A model that answers is easy. A model that acts needs scoped tools,
memory it can trust, and a governor that can refuse.
Then it has to survive a deploy.

$ uptime
7 years   systems -> cloud & devops -> agentic ai

$ echo $FOCUS
the engineering around the model, not the model

The stack

GenAI & Agentic

Bedrock AgentCore Strands LangGraph MCP A2A RAG Guardrails

Cloud & Infra

AWS Kubernetes Terraform Docker Jenkins Linux Ansible

Backend & Data

Python FastAPI Node SageMaker OpenSearch ServiceNow

Currently

+ NOW    agents that read an incident, ground themselves in the SOPs,
+        and propose a fix for a human to sign off
+ NOW    RAG scored for faithfulness and context recall, not eyeballed
+ BUILT  zero-egress inference: llama.cpp with local RAG on-box,
+        so the assistant runs where the data is not allowed to leave
The capability tree 
~/skills
├─ genai-agentic/
│   ├─ rag                retrieve, rerank, grade against source
│   ├─ agentic-rag        retrieval as a conditional tool call in a loop
│   ├─ bedrock            Knowledge Bases, Guardrails, model lifecycle
│   ├─ agentcore          Runtime, Gateway, Memory, Identity, Observability
│   ├─ frameworks         Strands, LangGraph, CrewAI, Agents SDK
│   ├─ mcp / a2a          write-once tool integration, agent-to-agent
│   ├─ memory             episodic, semantic, procedural
│   ├─ multi-agent        supervisor-worker, agent-as-tool, swarm, graph
│   ├─ rag-eval           RAGAS, faithfulness, context precision + recall
│   ├─ observability      Langfuse, Phoenix, OTel GenAI, P95 + cost
│   └─ guardrails-hitl    Bedrock + NeMo Guardrails, human gates
├─ cloud-infra/           AWS · Kubernetes · Terraform · Docker · CI/CD · SRE
├─ backend/               Python · FastAPI · Node · REST · ServiceNow
└─ data-mlops/            SageMaker · MLflow · vector at scale · pipelines
Laws I wrote for myself 
01  If it can't be rebuilt from a file, it doesn't exist.
02  The model is the easy part.
03  Anything irreversible waits for a human.
04  Write the fix as carefully as the exploit.
05  Boring, observable, repeatable. In that order.
06  If I'm the only one who understands it, it's broken.

Paper that agrees with me

Claude Certified Architect · Professional · AWS Solutions Architect · Associate · Azure DevOps Engineer · Agentic AI Engineer · Professional

Popular repositories Loading

  1. gamers-shop gamers-shop Public

    A Shopping Website build with Angular 4 and Firebase.

    TypeScript 2

  2. react-slack-chat react-slack-chat Public

    A chatting app made with React, Redux and Firebase

    JavaScript 1 1

  3. react-crud react-crud Public

    A CRUD diary app made with React Redux and Firebase.

    JavaScript

  4. sadiq-git sadiq-git Public

  5. autoscaler autoscaler Public

    Python

  6. jenkins-ai-remediation-plugin jenkins-ai-remediation-plugin Public

    Jenkins plugin that uses GenAI to diagnose failures and push fixes.

    Java