# Abhinav Bansal

The Tech Plumber
Wherever engineering leaks, I go in, then build the team and the tools so it stays fixed. Test automation, then modernization at scale, now AI agents.

Vice President at Morgan Stanley leading engineering teams through agentic AI adoption and tech modernization, after five years modernizing technology at scale across the firm's business units. Known as the Plumber: wherever something needs fixing, I go in. Strong Python engineer who builds multi-agent systems with LangChain, MCP and the Claude Agent SDK. I love building cookie-cutter tools that let every team start from a working pattern, and presenting them as a developer advocate. Came up through test automation and DevOps, and co-invented a patented test automation engine.

Location: Glasgow, Scotland, UK. Email: abhibansal60@gmail.com.
LinkedIn: https://www.linkedin.com/in/erabhinavbansal/
GitHub: https://github.com/abhibansal60

## Experience

### Vice President, AI Platform & Tech Modernization, Morgan Stanley (Dec 2024 to present), Glasgow, UK
Relocated to Glasgow to lead technology modernization under the firm's Refresh Tech strategy.
- Lead agentic AI adoption and literacy across engineering teams: developer advocacy programmes on agentic frameworks (LangChain, MCP, Claude Agent SDK) and LLM integration patterns
- Build specialised agents for development workflows and DevOps operations
- Drove Kubernetes and Docker adoption across development teams, with 30% faster deployment cycles
- Give architectural guidance on multi-agent orchestration, developer tooling and LLM integration

### Vice President, Morgan Stanley (Jan 2024 to Dec 2024), Bengaluru, India
- Led 5+ engineering teams through tech modernization, with a 40% improvement in deployment frequency
- Led the ACT Conference Portal, recognised with the India Technology Showcase Award
- Pioneered Kubernetes adoption with documentation and training that sped up adoption across the organization
- Set up a formal mentoring programme; 3 mentees are now in advanced technical roles

### Senior Manager, DevOps, Morgan Stanley (Jan 2021 to Jan 2024), Bengaluru, India
- Scaled the Engineer as a Service model: 35% faster modernization rollouts across business units
- Reusable architecture patterns that cut time-to-market for new applications by 28%
- DevOps and SRE practices with global teams: 45% fewer production incidents
- Frameworks and tools used across the firm, improving quality metrics by 20%
- Led the firm's Transformation network for DevOps practice adoption

### Manager, Test Automation, Morgan Stanley (Jan 2020 to Jan 2021), Bengaluru, India
- Built an email-embeddable Cucumber reporting tool for QA teams
- Built DevOps tooling and a worklog dashboard on the Bitbucket API, adopted by multiple teams

### Senior Associate, Test Automation, Morgan Stanley (Aug 2018 to Jan 2020), Bengaluru, India
- Led a generic, organization-wide test automation framework from concept to code

### System Engineer, Infosys (Aug 2016 to Aug 2018), Bengaluru, India
- Designed a UI automation framework in Java using the Reflection API
- Built a report generation module with data cleansing that made the team 11 times more productive
- Designed the frontend and REST API of the GST Refund Application
- Clients: Telstra and the Government of India

### Founder, DollarsGPT.com (May 2013 to Dec 2015)
Consumer rewards platform paying users in gift cards and cash for micro-tasks. Bootstrapped while at university.
- Grew from zero to 20,000 users with 1,500+ monthly actives through performance marketing and partner integrations
- Ran the whole product: platform development, partner onboarding, payments and growth

## Systems built at Morgan Stanley (internal, described only)
- Boardroom: A multi-agent orchestration system: three specialised agents and an orchestrator, built with LangChain and the Claude Agent SDK. It is the internal reference implementation for agentic AI at the firm.
- BLAZE (US patent, co-inventor): BDD Lean Automation Zenith Engine: turns API specifications into executable Java test suites using behaviour-driven development. Granted as US Patent 12,321,256 B1, "Behavior data driven automation test code generating system and method", on 3 June 2025, assigned to Morgan Stanley. https://patents.google.com/patent/US12321256B1/en
- Internal events and conference platform: Built and evolved over 5+ years into the standard for global internal events. Added AI features: proposal submission and review with a recommendation engine for session curation, content filtering, and a Hacker News-style knowledge platform with AI ranking.
- AI developer tooling: A VS Code extension for in-place codebase review and modernisation, CLI skills for DevOps automation, and reusable skills infrastructure that cut onboarding and deployment tasks from days to minutes.
- Pull request chat: A chat bubble on the pull request page so authors can talk to reviewers in place, saving 375+ person-days of effort a year.

## Open source projects
- tidy (https://github.com/abhibansal60/tidy): Cleans up Gmail and YouTube subscriptions with AI without letting AI loose on your account. A small model judges each email; plain, tested code decides what that judgment may do; nothing risky happens until you approve.
- ai-workbench (https://github.com/abhibansal60/ai-workbench): A Claude Code skill that sets up a matching matrix-green and cyber-purple dev environment on Linux: prompt, terminal, editor, diffs and status line. No sudo; it reads your config before touching it.
- awesome-copilot-toolkit (https://github.com/abhibansal60/awesome-copilot-toolkit): A VS Code extension to browse and install custom instructions, prompts and chat modes from the github/awesome-copilot repository without leaving the editor.
- macify (https://github.com/abhibansal60/macify): Makes Ubuntu GNOME look like macOS in one command, with a one-command revert.

## Skills
- Agentic AI: LangChain, MCP, Claude Agent SDK, Anthropic and OpenAI APIs, Pipecat, Deepgram, Twilio
- Platform: Docker, Kubernetes, AWS, Terraform, Jenkins, CI/CD
- Languages: Python, Java, TypeScript, React, SQL
- Testing: BDD, Cucumber, Selenium, TestNG
- Advocacy: Developer advocacy, Presenting, Cookie-cutter tooling, Mentoring

## Awards and certifications
- India Technology Showcase Award, Morgan Stanley. For the ACT Conference Portal.
- HashiCorp Certified: Terraform Associate (003)
- Site Reliability Engineering (SRE)

## Education
- Bachelor of Technology, Computer Science, Maharishi Markandeshwar University (2012 to 2016)

## Talk topics
- Getting agents past the change board: What it takes to move agentic AI from demo to production in a regulated enterprise: guardrails, literacy programmes, and the reference implementations that make teams trust it.
- Your test suite was the first agent harness: Ten years of BDD and test automation taught the same lesson agents need now: put deterministic checks around the part you don't trust.
- Cheap judgments, expensive builders: Splitting work between a frontier model that writes the code once and a small model that makes thousands of decisions, with real cost numbers from tidy.