Backend systems,
built with intent.
I'm Abishek — a CS student building distributed systems, financial ledgers, and the quiet infrastructure that has to hold up under load, failure, and review.
Built with a backend-first stack
Python
Golang
Java
CSS
C++
C
JavaScript
HTML
Distributed Systems
Research
PostgreSQL
LLM
Python
Golang
Java
CSS
C++
C
JavaScript
HTML
Distributed Systems
Research
PostgreSQL
LLM
Systems you can explain, test, and trust.
I care about systems that can be explained, tested, and trusted. That shows up as release governance, transactional storage, idempotent ledgers, distributed coordination, and explicit failure handling.
Alongside backend engineering, I explore research-heavy problems in AI robustness and contribute to open source inside real, production-scale codebases.
- VenturesReleaseGate · AgentLens
- SystemsShardDB, MIDAS Ledger, Job Orchestrator
- ResearchAI safety, failure modes, reliability
- Open sourcePRs across OpenSearch, Bazel, Kubernetes, FastAPI
- GoalBackend / platform internships where depth matters
Things I'm building.
Two products I founded — release governance and AI-agent observability.
Founder · 2025—
ReleaseGate
Evidence infrastructure for software changes — sign every deploy, prove every change.
Declarative policy enforcement, immutable audit snapshots, offline-verifiable proof packs, and a customer-installable GitHub Action that signs and verifies deploys.
Founder · 2026—
AgentLens
When your AI agent breaks, AgentLens tells you which decision caused it — and what to change.
Decision-level observability for AI agents. Instead of scrolling raw traces, you get the exact decision behind a failure and a concrete fix. Works with Anthropic, OpenAI, LangGraph, CrewAI, and AutoGen.
Things I've shipped.
A few projects where the hard part was correctness under concurrency and failure.
ReleaseGate
Evidence infrastructure for software changes — sign every deploy, prove every change. Declarative policy enforcement, immutable audit snapshots, and offline-verifiable proof packs.
AgentLens
Decision-level observability for AI agents — when an agent breaks, AgentLens pinpoints exactly which decision caused it and what to change. Works with Anthropic, OpenAI, LangGraph, CrewAI, and AutoGen.
Adaptive Release Controller
Self-adaptive CI/CD system that uses risk modeling and feedback loops to dynamically decide deploy, canary, or block actions from real-time signals and historical outcomes.
Change Risk Predictor
ML pipeline that scores the deployment risk of incoming code changes from commit history, CI signals, and ownership patterns.
ShardDB
Fault-tolerant transactional key-value store with custom Raft, 2PC, snapshots, and chaos-tested linearizability checks.
MIDAS Ledger
Concurrency-safe financial ledger with double-entry accounting, idempotent transfers, and invariant-based stress testing.
Job Orchestrator
Multi-tenant job orchestration system with race-free leasing, fairness controls, transactional outbox delivery, and observability.
Travel Buddies
Social travel platform for finding travel companions.
StorePro Fullstack
Fullstack e-commerce application with React, Node.js, and MongoDB.
Where I've worked.
Performing Arts Center Staff
Stockton University
Supported operations at the Performing Arts Center.
Resident Assistant
Stockton University
Fostering inclusive environments and managing crisis situations with composure.
Math Tutor
Stockton University
Tutored students in mathematics, helping them understand complex concepts.
Where I'm learning.
Stockton University
B.S. Computer Science
Building depth in backend engineering, distributed systems, and applied AI through projects, research, and open-source work.
Summer Entrepreneurship and Business Academy (SEBA)
Stockton University
Participated in and contributed to the entrepreneurship academy.
Good to know.
A few quick answers if you're considering reaching out.
Let's build something that actually matters.
Open to backend and platform internships, research collaborations, and good engineering conversations.