Academic Companion
A desktop class manager for students. Drop in a syllabus and it pulls out deadlines and exams. Record a lecture and it transcribes and summarizes it. It can run fully local with Ollama or use Groq in the cloud.

I'm Sahil, a final-year computer science student at RIT, and I build AI agents and the apps around them. One of them phones vendors and gets you quotes for your event. Another writes, edits and posts my Instagram reels. Another keeps me honest about my habits. I build because I want to solve real problems for real people, not just because a technology is interesting on its own. So I ship things, put them online and keep fixing them.
This summer I did machine learning research at Clarkson University, training one model to recognize people by the way they move a mouse. I presented it at RAPS, Clarkson's research showcase, and won Best Undergraduate Oral Presentation in my category. The work continues this fall as an independent study at RIT.
I graduate in December 2026. I'm looking for full-time software or AI engineering roles starting January 2027, and co-ops or internships in that window too. I'm based in Rochester and happy to relocate anywhere in the US, with New York City at the top of the list.
I start with a problem, get a working version out fast, then keep going until it holds up with real use. Coding agents like Claude Code are part of my daily toolkit, and in a lot of my projects they are part of the product too. Before RIT I studied at VIT in India, where I ran sponsorships and helped put on hackathons for Google Developer Student Clubs.
Voice AI · New
Dispatch finds local vendors for your event, calls and emails them, and brings back real quotes.
Planning an event means a lot of phone calls. Venues, caterers, florists, DJs, photographers. Every one of them wants to talk, plenty of them don't pick up, and every quote comes back in a different format. Dispatch does that part for you. You pick the event type and the services you need, describe the event, and it goes and gets the quotes.
Under the hood it finds local businesses with the Google Places API, then contacts each one the way that category actually works. Some vendors live on the phone, so a Retell AI voice agent calls them. Others run on email, so Resend sends a proper message from its own reply address. Call transcripts and email replies go through an LLM on OpenRouter, which turns them into structured quotes on your dashboard.
Continued on Page B2
Developer Tools · New
FrameWright is a marketplace of Remotion video templates that coding agents drive by writing JSON.
AI video tools usually mean a prompt box, a render queue and a bill for every render. I wanted something different for developers who already work in Claude Code or Cursor. FrameWright sells template packs, which are closed design systems for video. Your agent reads the pack's SKILL.md, fills in a JSON manifest, and the template handles all the motion and timing. The MP4 renders locally, so there's no per-render fee.
Continued on Page B3
Agentic AI
LifeOS is my personal dashboard of small apps on one Supabase database, with an always-on AI agent that talks to me on Telegram.
A personal operating system: small apps for habits, workouts, diet, budget and more, all on one Supabase database. A daemon on an old laptop runs an AI accountability agent over Telegram, and it can build new apps when I describe them.
Continued on Page B4
Creator Tools
autoreel turns one talking-head clip into a finished, edited Instagram Reel and posts it.
The daily Reels pipeline behind @21stcentury.sahil. It finds topics, drafts scripts in my voice, edits my clip into a reel with Remotion, posts it and sends tool links to commenters. I run all of it from one Telegram chat.
Continued on Page B5
Research
T&C Applied rewrites privacy policies into plain English, flags unfair clauses and answers follow-up questions.
A privacy-policy simplifier built on my own readability study. Paste a policy and it rewrites it at an 8th-grade level or below, sorts clauses into five categories, flags biased terms and lets you ask questions about it.
Continued on Page B6
Desktop
Academic Companion turns syllabi, assignment PDFs and recorded lectures into deadlines, summaries and a chat over everything in a class.
A desktop class manager for students. Drop in a syllabus and it pulls out deadlines and exams. Record a lecture and it transcribes and summarizes it. It can run fully local with Ollama or use Groq in the cloud.
Continued on Page B7
Competitive Coding
BeatCode tests whether developers can actually ship with AI coding agents, then rates them with Elo.
A competitive coding platform where you solve build and fix challenges with AI agents inside cloud sandboxes. A trusted grader scores the result and updates your Elo. Version 1 works, and version 2 is in progress.
Continued on Page B8
Browser Agents
JobOps finds jobs, tailors a resume for each one and pre-fills the application, then stops and waits for me.
A job-search command center that runs on my laptop. It pulls postings from five sources, scores them with headless Claude, tailors a LaTeX resume per job and pre-fills applications in Chrome. It never submits. I review and send every one.
Continued on Page B9
Mobile AI
A phone homescreen where every icon is a mini-app that Claude wrote and a build server compiled from a plain-English description.
Describe a tool, Claude writes it as React code, a build server bundles it, and it runs on your phone in a sandbox. Each mini-app gets storage, the mic, the calendar, the camera and AI through a native bridge.
Continued on Page B10
All of it on one shelf: The Archive, Section B

Science
I spent the summer teaching one model to recognize people by how they move a mouse. It won an award. It also has a flaw I caught myself, and fixing it is my fall project.
The idea is simple. Everyone moves a mouse a little differently: how fast you go, how you curve toward a button, how long you pause before you click. If a computer can learn that, it can tell whether the person using it is really you. Basically FaceID, but for your mouse.
I spent Summer 2026 at Clarkson University as an undergraduate researcher in Prof. Daqing Hou's group, mentored by Charles Devlen. The work I built on trained a separate model for every user in every dataset. Across three public datasets, that came to 59 models. It works, but it does not scale, and it cannot recognize anyone it was not trained on.
So I built one model for everyone. It is a 2-layer BiLSTM in PyTorch that reads a stretch of mouse movement and turns it into a 128-number embedding, a kind of fingerprint for how you move. I trained it with semi-hard triplet loss, the same idea behind FaceNet and TypeNet: pull samples from the same person together and push different people apart.
Continued on Page D1
Machine Learning Researcher (REU) · Potsdam, NY
Clarkson is a research university in northern New York, and I spent the summer in Prof. Daqing Hou's group working on behavioral biometrics.
Security Intern · Sunnyvale, CA (hybrid)
Swif.ai builds a device management (MDM) platform that companies use to manage and secure their devices.
App Developer Intern · Santa Clara, CA
Protego makes an Android app that detects scam messages.
App Developer Intern · Bengaluru, India
Setu is a fintech company that builds APIs other developers use to add financial services to their apps.
Full listings on Page C1
Tech I actually use. Number = projects, jobs and coursework that use it.
| Sym | Name | Used | Chg |
|---|---|---|---|
| TS | TypeScript | 19 | ▲ |
| PY | Python | 11 | ▲ |
| RUST | Rust | 2 | ▬ |
| JAVA | Java | 3 | ▬ |
| CPP | C/C++ | 2 | ▬ |
| Sym | Name | Used | Chg |
|---|---|---|---|
| NEXT | React / Next.js | 18 | ▲ |
| NODE | Node.js | 8 | ▲ |
| TRCH | PyTorch | 1 | ▲ |
| TAUR | Tauri | 2 | ▬ |
| EXPO | React Native / Expo | 3 | ▬ |
| RMTN | Remotion | 3 | ▲ |
| ASTR | Astro | 1 | ▲ |
| Sym | Name | Used | Chg |
|---|---|---|---|
| CLDE | Claude Code & agent orchestration | 9 | ▲ |
| RAG | RAG | 4 | ▬ |
| LLM | Local LLMs (Ollama, on-device) | 5 | ▬ |
| VOIC | Voice agents (Retell, Vapi) | 1 | ▲ |
| WSPR | Whisper | 5 | ▲ |
| Sym | Name | Used | Chg |
|---|---|---|---|
| SUPA | Supabase / Postgres | 8 | ▲ |
| DOCK | Docker | 3 | ▲ |
| EDGE | Cloudflare / Vercel | 7 | ▲ |
| PWRT | Playwright | 4 | ▲ |
| SQL | SQLite | 6 | ▬ |
More on Page C4
Situation wanted
I'm looking for full-time software or AI engineering roles starting January 2027, and I'll move anywhere in the US (New York City first).
Resume
Education, experience, projects and skills on a single sheet. PDF, prints nicely.
Download (PDF)Code
46 repositories and 579 contributions in the last year. The big ones are private, but ask and I'll walk you through them.
github.com/SahilBachu ↗Now showing
A tech page on Instagram. The reels are made by my own pipeline, autoreel.
instagram.com/21stcentury.sahil ↗Languages spoken
Personals
Based in
Happy to relocate anywhere in the US. New York City would make my year.