Hi,
I'm Akshat Bindal.
I convince messy data to behave, and build AI agents for fun.
TL;DR
I basically move massive piles of data around the cloud and make sure none of it catches fire. 3+ years of building data pipelines by day, and tinkering with fun AI agents by night. If a task is boring, I automate it. If it's cool, I build it.
or scroll down (I recommend the latter one).
What I do: 9 to 5
Senior Analyst – Credit & Fraud Risk
Playing digital undertaker and AI detective for enterprise credit risk. Architected automated archival pipelines so petabytes of analytical data age gracefully (and legally), while unleashing BigQuery and GenAI to sniff out sensitive data like an ultra-diligent compliance hound.
Analyst – Credit & Fraud Risk
Waged an all-out war on latency and manual toil. Banished sluggish on-prem Hive and PySpark behemoths into lightning-fast BigQuery SQL (slashing runtimes by 80%), built Cloud Composer pipelines that reclaimed 4 days of monthly sanity, and automated internal workflows so humans could stop doing robot chores.
Associate Consultant
Engineered in the high-stakes boiler room of global banking on Oracle FLEXCUBE. Wrote finely tuned Java and battle-hardened PL/SQL where a microsecond delay or a misplaced decimal isn't just a bug—it’s an existential crisis for millions of daily transactions.
Intern – Robotics Tech Team
Basically a robot whisperer. Built an IoT mobile framework to monitor industrial robots from a phone, then crunched sensor telemetry to cure heavy arc-welding machines of their violent tremors—slashing vibration anomalies by 60%.
What I do: 5 to 9
Ensuring I thrive in a post-AI world by actively learning and building with AI. I develop AI agents to solve real-world problems—or construct problems to fit the solutions I invent. I simply love engineering cool concepts into reality.
What I'm Good At
- Architecting Massive Cloud Systems Deep, end-to-end mastery across the Google Cloud ecosystem. Backed by official credentials as a Google Cloud Certified Professional Cloud Architect [I don't believe you ] .
- Building with AI Agents Autonomous workflows, prompt architecture, MCP servers, and LLM evaluation. Frameworks come and go every week, but the intuition to build working, resilient AI systems stays.
- Hyper-Speed Execution & Prototyping Turning vague ideas into working prototypes and production-grade tools at 10x speed using bleeding-edge AI dev workflows.
What I've made
Style Vision
Because computer vision algorithms should have better fashion taste than my friends.
Jules MCP Server
A Model Context Protocol server that turns Google's Jules into an AI coding powerhouse that actually follows instructions.
Vibe Trips
AI-crafted travel itineraries based on pure vibes, so you never end up trapped in another tourist scam.
Finance MCP Server
Feeding live financial datasets directly to LLMs through standard protocols without letting them hallucinate stock numbers.
Network Security DS
Teaching machine learning models to spot sketchy enterprise network behavior before the security team's alarm goes off.
Global Disaster Monitor
An automated cloud pipeline tracking worldwide crisis events in real time—because disaster data waits for no one.
Where I turned into an Engineer
National Institute of Technology Kurukshetra
CGPA: 9.38