Skip to content
View dhrub-git's full-sized avatar
🏠
Working from home
🏠
Working from home

Highlights

  • Pro

Block or report dhrub-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.

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
dhrub-git/README.md
Dhrubajyoti Biswas - Principal Generative AI and Cloud Architect

Website · Credentials · Projects

I am an engineering and architecture leader with 18+ years of experience turning ambiguous enterprise problems into working systems. I combine executive discovery and enterprise architecture with hands-on delivery of agentic AI, RAG, cloud-native platforms, and repeatable adoption playbooks.

My experience spans public sector, SaaS, transportation, financial services, automotive, and technology.

Impact at a glance

Experience Team leadership Automation Operations
18+ years 12-person team 95%+ 25% lower
Engineering and architecture Agentic RAG delivery RCA report generation Mean time to resolution

Additional outcomes include 40% less manual support intervention and 30% fewer customer escalations and reopened requests through AI-assisted triage and knowledge retrieval.

Selected systems

Project What it demonstrates
construct.da
Live demonstration
Evidence-backed development-approval screening for Australian residential projects. Document ingestion, OCR, spatial data, asynchronous workflows, and advisory reporting in a full-stack TypeScript application.
Mandate Agentic AI assurance and policy generation with guided onboarding, LangGraph orchestration, web research, streaming progress, and deterministic EU AI Act risk classification.
Intelligent Briefing Notes Architecture and implementation reference for durable, multi-party approval workflows using Temporal, including state transitions, retries, observability, testing, and operational scaling.
AI Security Practical knowledge base covering AI governance, privacy, adversarial ML, model risk, security controls, and assurance frameworks.

Leadership and engineering focus

  • Forward-deployed AI: customer discovery, workflow mapping, rapid prototyping, technical validation, and production handover
  • Reliable agentic systems: RAG, tool use, multi-agent orchestration, evaluation, guardrails, observability, and human review
  • AI governance and assurance: converting policies, risk frameworks, and regulatory requirements into practical engineering controls
  • Cloud and platform architecture: OCI, AWS, Azure, Kubernetes, serverless, APIs, integrations, and event-driven systems
  • Engineering leadership: creating clarity, raising technical standards, and enabling teams to deliver through ambiguity

Technology landscape

Python TypeScript JavaScript Java Go SQL React Next.js Spring Boot LangGraph LangChain Temporal PostgreSQL Kubernetes OCI AWS Azure

Operating principles

  1. Start with the user workflow and a measurable outcome.
  2. Make assumptions, constraints, and failure modes explicit.
  3. Build the smallest working system that can prove or disprove value.
  4. Instrument quality, reliability, cost, and adoption.
  5. Turn repeated delivery patterns into capabilities other teams can reuse.

Enterprise AI · Forward-Deployed Engineering · Architecture · Delivery Leadership

Pinned Loading

  1. karpathy-nanochat karpathy-nanochat Public

    Personal study fork of Andrej Karpathy's nanochat, with added learning notes. All code (c) Andrej Karpathy (MIT).

    Python

  2. ai-security ai-security Public

    Understanding AI Security - Preparing for CISSP, CISM, CISA, and AIGP certifications. Comprehensive documentation, examples, notebooks, and research on AI security practices.

    1

  3. mandate.sh mandate.sh Public

    AI Assurance and Policy Generation - Mandate is a Turbo monorepo for generating AI governance policies from a guided company onboarding flow

    TypeScript 1

  4. briefing-notes briefing-notes Public

    Intelligent Briefing Notes

    1

  5. awesome-llm-apps awesome-llm-apps Public

    Forked from Shubhamsaboo/awesome-llm-apps

    LLM apps with AI Agents and RAG using OpenAI, Anthropic, Gemini and opensource models.

    Python

  6. construct.da construct.da Public

    construct.da — Codex Hackathon DA approval worktree by Dhrub Biswas

    TypeScript 1