Chicago, IL · open to SWE & AI engineering roles

Hi, I’m Sriya.I build AI agents, and the products around them.

Multi-step agents that reason, call tools, and account for every decision, grounded in real data and measured rather than assumed.

View my resume Get in touch

Scroll

01About

I care more about whether a system makes the right call than whether it sounds smart making it.

I finished my M.S. in Computer Science at Indiana University Bloomington in May 2026. Most of my time goes into AI agents: the kind that work through a question rather than just answering it, calling tools, checking real data, deciding what’s next, and keeping a record of why. I build both halves, the reasoning and the product around it.

Based in Chicago, IL, and happy to relocate

M.S. Computer Science, Indiana University Bloomington, May 2026

Building agents that reason, use tools, and make real-time calls

projects shipped & documented

cloud & platform certifications

roles across data, backend & consulting

02Experience

Every role so far has had the same shape: something was manual and confusing, so I automated it and made it make sense.

Jun 2026 – Present

Senior Consultant

Heartland Community Network

Indiana, US · Remote

01I map where work actually gets stuck across 5 business processes, and turn that into a technical roadmap for automation rather than a wishlist.
02I prototype in Python and SQL before anyone commits engineering time, which has saved a lot of rework that would otherwise have surfaced much later.

Aug 2025 – May 2026

Inventory Data Analyst

Indiana University Dining

Bloomington, IN

01Dining halls were over-ordering because nobody had a clear read on demand. I built automated inventory and demand-trend pipelines across 5+ locations, and overstock dropped.
02Every team tracked things differently, so the numbers never lined up. I rebuilt the ETL pipeline and set validation and logging standards across 3+ workflows.
03The people reading these reports weren’t technical, so I designed the SQL for clarity over cleverness. Nobody needed training, which is why they actually got used.

Jul 2023 – Jul 2024

Software Engineer

Next Page Technologies

Hyderabad, India

01I owned 20+ modules of a CRM a client team used daily, from the database schema up through the REST APIs running their lead and sales workflows.
02Automated what used to be manual inside it: rule-based processing, multi-step approvals, notifications, validation. The same steps ran the same way every time.
03Built role-based, stage-aware access control that held consistently across frontend and backend, not just in one place.

Jan 2022 – May 2022

Robotic Process Automation Intern

AICTE-EduSkills

Hyderabad, India

01Replaced manual copying from invoices and web forms with Python extraction and validation pipelines. Cut processing time by about 25%.
02Built four reusable automation tools rather than one-off scripts, with REST notification and reporting so results reached stakeholders without anyone chasing them.

03Selected work

Five systems, five different questions about what “autonomous” should actually mean.

Aug 2026

NoteKit

Turns any article or PDF into a study guide you can actually trust

Hand it a web page, a PDF, or your own notes and it returns a study module with a quiz, where every fact links back to where it came from, so you’re never studying something the model made up. Hybrid retrieval underneath: semantic and keyword search, fused, then reranked.

PythonFastAPIPostgreSQLpgvectorNext.js

95.3%

of claims traced to the source

16/16

times it refused instead of guessing

65%

faster first draft (52s → 18s)

48%

cheaper per course

View on GitHub

May 2026

RatePilot

An AI that haggles your bills down so you don’t have to

The bill-negotiation call you keep putting off, made by an agent instead. It looks up what competitors actually charge, works out an angle, and emails the provider. When they reply it decides on its own whether to accept, counter, or hand off to you, logging why every time.

PythonLLM APIsTool UseGmail APIFastAPI + React/Vite

times I had to step in

moves it can make each round

100%

of decisions logged with a reason

View on GitHub

Sep 2025

Real-Time Edge AI Vision

A traffic camera that clocks your speed without touching the cloud

A Raspberry Pi watches a road, spots vehicles, and works out how fast they’re going. All of it runs on the board, with nothing sent to a server. YOLOv8 detects, tracking across frames estimates speed, and anything over the limit gets logged and flagged.

PythonYOLOv8Raspberry PiOpenCV

40%

faster inference on-device

<1s

to detect and log a vehicle

bytes sent to the cloud

View on GitHub

Apr 2026

Distributed Graph Processing Engine

When is a cluster worth it? I measured the exact line.

Does throwing more machines at a graph problem actually make it faster? I ran PageRank and shortest-path across 5M-edge graphs on a 5-node cluster, 300+ runs over every worker and partitioning combination. On big graphs the cluster won by 3.3×. Once the data fit in one machine’s RAM, coordination overhead made it 25–400× slower.

Apache SparkGraphXScalaPythonDockerNSF Jetstream2

300+

benchmark runs across the sweep

3.3×

speedup on large graphs

25–400×

overhead once data fits in RAM

31.7%

less wall-time from locality tuning

View on GitHub

Aug 2026In progress

Delta

Teaching an AI to rewrite its own instructions, carefully

An experiment in letting an AI rewrite its own instructions without taking its word for it. One agent writes SQL from plain English; a second watches where it fails and rewrites the prompt. The rewrite only survives if it beats the old one on a held-out test, with a confidence interval attached.

PythonLLM APIsStatistical TestingSpider Benchmark

100

test questions evaluated so far

methods it’s being benchmarked against

View on GitHub

04Toolkit

What I reach for, from schema design to shipping the agent.

Filter by category, or hover a skill to see which project used it.

Languages

Backend

Frontend

AI & ML

Data & Databases

Cloud & DevOps

Practices

05Education & credentials

Two degrees, five certifications, and one paper that made it into a conference.

Education

Indiana University Bloomington

M.S. in Computer Science

GPA 3.67 / 4.0 · Aug 2024 – May 2026

Applied Algorithms · Machine Learning · Computer Vision · Data Mining · Computer Networks · Distributed Systems · Cloud Computing

KL University

B.Tech in Computer Science Engineering

GPA 8.97 / 10 · Aug 2020 – May 2024

Database Management Systems · Design & Analysis of Algorithms · Operating Systems · Object-Oriented Programming · Cybersecurity · Edge Computing

Publication

Survey of Non-English Language Compilers: Exploring the Diversity of Programming Languages

9th International Conference on Advanced Computing and Communication Systems (ICACCS), Coimbatore, India, 2023, pp. 1187–1193

Certifications

AWS Certified Solutions Architect – Associate

AWS Certified Cloud Practitioner (CLF-C02)

Microsoft Azure Fundamentals (AZ-900)

Red Hat Certified Enterprise Application Developer

Google Cloud Associate Cloud Engineer

06Contact

Building something worth reasoning about?

I’m open to full-time SWE and AI agent engineering roles. The fastest way to reach me is email, and I read every one.

sriyasahu.2003@gmail.com

GitHub· LinkedIn