Software engineer · Researcher

Jigyasa Hasija

Secure Multi-Agent Systems · Agent Security · Distributed Systems

M.S. Computer Science, Georgia Tech · 2026-2028

Jigyasa Hasija

I am a software engineer and researcher focused on making autonomous multi-agent systems reliable, secure, and measurable. My current research at Georgia Tech's RAIL Lab explores capacitated many-to-many multi-agent pickup and delivery: scalable task allocation and collision-free routing under payload constraints.

I am building toward secure multi-agent systems that coordinate through controlled tools, respect permissions and operational constraints, and remain resilient to failures or adversarial inputs.


Currently Working On

Current research

Capacitated Multi-Agent Pickup and Delivery

RAIL Lab · Georgia Tech

Building the first step of a broader MAPD research program: a reproducible benchmark for payload-constrained multi-robot logistics. Current work focuses on deterministic task traces, invariant-driven simulation, fair TPMC/RMCA baselines, and conflict-free routing with reservation-table space-time A*; the next phase will develop and evaluate a many-to-many pickup-and-delivery algorithm on this foundation.

In progress · Future build direction

Secure Multi-Agent Incident Response Sandbox

Current build direction

Developing a simulated incident-response environment for permissioned agent coordination, tool safety, fault tolerance, and adversarial evaluation.


Research

Reproducible Benchmarking for Capacitated Multi-Agent Pickup and Delivery

Current research · RAIL Lab · Georgia Tech

I am building a reproducible framework for capacitated many-to-many multi-agent pickup and delivery (MAPD) in payload-constrained logistics environments, with identical task traces, maps, starting states, horizons, and seeds for fair comparison.

Benchmarking is the first phase, not the endpoint: a frozen runnable baseline, explicit Task, Agent, ScheduleEvent, and WorldState models, replayable event logs, and invariant checks for capacity, ordering, assignment, conflicts, deadlines, and executed cost.

After validation, I will develop and evaluate a many-to-many MAPD algorithm. Initial work uses TPMC and RMCA under capacity 1, with A* estimates and reservation-table space-time A*; later experiments will study higher-capacity insertion, allocation, routing, throughput, planning time, and execution cost.

Research methods

  • Frozen runnable repository baseline
  • Replayable task traces and identical seeds
  • Invariant-driven validation
  • A* shortest-path estimates
  • Reservation-table space-time A*
  • TPMC and RMCA initial baselines
  • Independent event-log replay
  • Multi-seed evaluation
  • Algorithm development after benchmark validation

FPGA Implementation of Residue Number System based Digital Differentiator

Prior Research · Advised by Prof. Tarun Kumar Rawat · Netaji Subhas Institute of Technology

  • The Problem: Traditional binary arithmetic introduces severe computational carry-propagation delays and power overheads when executing high-frequency digital signal processing on resource-constrained or edge devices.
  • The Approach: Evaluated alternative mathematical frameworks by designing and implementing a full-band Digital Differentiator using a three-moduli Residue Number System (RNS). Built the underlying hardware description architectures in Verilog HDL and simulated execution workflows on an Artix-7 FPGA target using the Xilinx Vivado design suite.
  • The Outcome: Demonstrated that custom-tailored RNS modular structures compress arithmetic bottlenecks and decrease total processing time for edge digital systems compared to standard infinite impulse response configurations.

Conference Presentation: “FPGA Implementation of Residue Number System based Digital Differentiator” — peer-reviewed and presented at The International Conference on Signal Processing, VLSI and Communication Engineering (ICSPVCE), Delhi Technological University.


Selected Engineering & Systems Projects

LLM-Assisted Test Automation and MCP Tool Reliability

Software Development Engineer · Veeam

MCP · TypeScript · Playwright · LLM tool use · Failure analysis

  • Architected proof-of-concepts exploring the integration of Anthropic's Model Context Protocol (MCP) with TypeScript-based browser automation engines.
  • Evaluated the reliability and failure boundaries of frontier models (Claude) in autonomously orchestrating and generating complex end-to-end UI verification flows from high-level natural language intent, optimizing tool-calling efficiency.

Multi-Tenant Systems Reliability and Zero-Trust Isolation

Software Development Engineer · Microsoft Security Organization

Distributed systems · Cloud security · Tenant isolation · Reliability

  • Authored a technical design paper evaluating distributed caching strategies for a compliance-management service to mitigate heavy API latency bottlenecks.
  • Re-engineered critical backend services and multi-tenant distributed endpoints using .NET and Azure, reducing p95 latency by up to 5 seconds while sustaining a 98%+ system reliability rate.
  • Remediated high-impact cloud security vulnerabilities under a company-wide secure future initiative, analyzing data boundaries to enforce robust data protection and isolation standards.

Low-Overhead Infrastructure Provisioning & Automation Pipelines

Engineer · Samsung R&D Institute

Focus: Infrastructure as Code (IaC) · Pipeline Efficiency · Resource Optimization

  • Audited multi-tier system environments to eliminate deployment constraints and optimize manual provisioning overhead.
  • Programmed automated cloud infrastructure blueprints using the AWS Cloud Development Kit (CDK), cutting environment setup time latency by 80%.
  • Overhauled application delivery mechanisms by engineering CI/CD automation pipelines for serverless AWS Lambda functions, decreasing deployment cycle times by 75%.

Skills

Demonstrated in production/research, with current research and development focus clearly marked.

Agentic Systems

  • Model Context Protocol (MCP)
  • TypeScript
  • Playwright automation
  • LLM tool use
  • Agent workflow reliability

Current Research and Development Focus

Current research/development focus

  • Multi-agent task allocation
  • Capacitated MAPD
  • Collision-free routing
  • Agent permissions
  • Least-privilege tool access
  • Adversarial evaluation
  • Fault tolerance

Distributed Systems and Backend

  • C#
  • .NET
  • Java
  • Python
  • Spring Boot
  • REST APIs
  • SQL
  • KQL
  • Microservices

Cloud, Reliability, and Security

  • Azure Durable Functions
  • Cosmos DB
  • Kusto
  • AWS Lambda
  • AWS CDK
  • CloudWatch
  • Jenkins
  • CI/CD
  • Multi-tenant isolation
  • Zero-trust data boundaries

Hardware and Prior Research

  • Verilog HDL
  • Xilinx Vivado
  • System Generator for DSP
  • MATLAB
  • Embedded C
  • FPGA systems

Beyond the Lab

🩰 Creative Discipline & Resilience

I am a trained Jazz dancer. I approach system architectures much like choreography — both demand precise timing, physical or computational constraints, and structural alignment. When forced into long periods of physical stillness due to three separate ankle fractures, I used that time to independently master advanced data structures and transition my technical focus from hardware engineering to software systems.

🗺️ Languages & Curiosity

I like picking up languages in my spare time. I learned German a couple of years ago, and I'm currently working my way through Italian, on and off, whenever I can steal a few minutes here and there. Alongside English and Hindi, it's another way I get to explore new cultures and ways of thinking.

📚 When My Monitor is Off (My Reading Stack)

I read a lot — usually tracking huge yearly goal across fiction, essays, memoirs, and science. You can take a peek at my full shelf of over 500 books on Goodreads, but here is a quick snapshot of what's currently on my desk:

Curiously flipping through right now

  • Letters from New York by Helene Hanff.
  • The Guardian and the Thief by Megha Majumdar.
  • The Hare with Amber Eyes by Edmund de Waal (A fascinatingly detailed trace of a hidden inheritance across generations)
  • A Brief History of Black Holes by Becky Smethurst, because sometimes software systems aren't complex enough, so I check in on space-time singularities.

A few all-time favorites

I'm drawn to rich world-building, multi-generational sagas, sharp prose, and books about books. A few of my favorites:

  • Classics: Emma by Jane Austen · Madonna in a Fur Coat by Sabahattin Ali
  • Literary fiction: All Passion Spent by Vita Sackville-West · East of Eden by John Steinbeck
  • Non-fiction: Mother Mary Comes to Me by Arundhati Roy · Dead and Alive (essays) by Zadie Smith
  • Books about books: Diary of a Bookseller by Shaun Bythell · The Girl Who Ate Books by Nilanjana Roy

Why this matters to me

“I am fascinated to presume, as a reader, that many types of people, strange to me in life, might be revealed, through the intimate space of fiction, to have griefs not unlike my own. And so I read.”

— Zadie Smith

That line resonates with me deeply. Reading is how I travel to places I could never go - Rome in the 1970s, quiet UK villages of the 1800s. It's how I come to know about fascinating things: antiquarian books, art, black holes, bookshops across the world, and endless new ideas. Every book leaves me a little more curious than it found me. And that, more than anything, is something I can't replace.