Verbatim
Verbatim is an intelligent platform that takes any video, summarizes it for quicker consumption, translates it into multiple languages, and then recreates the speaker’s lip movements to match the new audio—delivering a seamless, localized experience.
Origin and what it does
We built Verbatim at BoilerMake — a video understanding platform that goes beyond simple transcription. We didn't win, but our frustration turned into determination, and we took our hackathon project further. Verbatim translates speech in videos into 20+ languages, summarizes content for quick insights, creates a digital clone that lip syncs perfectly in the target language in the original speaker's voice, and uses memory level video embeddings for real time Q&A and topic search. The team was Pranav Neti, Sonny Chen, and Cindy Yang, who also coined the name.
Why we built it
Our goal was to fix dubbing, because too often, people who speak different languages hear dubs that don't match the tone or lip movement of the speaker. It creates an uncanny, unpleasant experience for viewers.
Why we kept going past the hackathon, and why it died
We wanted to see if we could take it further. We got pretty far in our work, but we quickly realized that the use case for international creators already existed on a large scale at production grade. We couldn't figure out a differentiator, and so the idea died out shortly thereafter.
What I'd do differently
The biggest thing that didn't work out well was that we didn't look for a real use case before building out the platform. We didn't clearly articulate the benefit to the customer (which may be why we didn't win) and also didn't realize that there was already a production grade product that existed on the market. The rule now, before I start building anything: Who needs this? What problem does it solve? What already exists that is solving this problem?
The technical bet I'd defend
Something incredible that we did was to parallelize the voice cloning and translation, followed by splitting the video into segments and performing lip sync on all segments with the translated speech in parallel, before combining them at the end. It reduced wait times by a TON, and since it was server side, the user wasn't affected by low latency.
For startups and founding teams
Karthik Thyagarajan is a founder-engineer studying Computer Science and Artificial Intelligence at Purdue. At Samsung Research America, he co-architected an always-on AI agent runtime for XR glasses. He is currently building Tavi, a voice teammate for engineering standups, and co-founded buildpurdue, a founder community at Purdue. His work spans agent infrastructure, on-device AI, product engineering, and evaluation.
For an early team building AI products, his strongest demonstrated fit is owning a broad technical problem from product decisions through implementation and evaluation. His current interests are live voice and video AI, AI-first wearables, and infrastructure that lets agents discover and use internet services.
Substantial ownership of agent systems
At Samsung Research America (May to August 2026), he co-architected an always-on agent orchestrator in Kotlin and owned half of a 25+ module runtime. His reported on-device router achieved 95% signal suppression and 90% intent-routing accuracy. His work also covered tool permissions, model routing, memory, and phone control, connecting model behavior to a usable device experience.
Builds teams and operating systems around a problem
As buildpurdue's co-founder and co-president, he helped turn an informal gathering into a founder community with a selective cohort, weekly co-working, funding, and founder-led problem-solving workshops. He built much of its web platform and integrated CRM, applications, events, and check-ins. His account describes data on around 500 students and a mailing list of about 1,000.
Makes technical claims testable
He built PARM, an open-source causal benchmark for when an agent should use personal memory: 157 frozen scenarios and 471 executable cases, with deterministic scoring and versioned replay artifacts. Its portfolio account explicitly distinguishes development results from held-out claims. He also built google-tools-mcp, a published MCP server exposing 169 Google Workspace tools through one OAuth flow.
Pursues difficult product problems
He is currently building Tavi, a voice teammate intended to answer codebase questions and handle small action items during engineering standups. The project tackles multi-person turn-taking, when to stay quiet, sandboxed coding, and synthetic meeting evaluations. This is ongoing work, rather than a claim of shipped adoption. His published interests explain the direction behind it.
Co-founder who ships and thinks about distribution
At Repple (November 2025 to June 2026), he co-founded a competitive fitness app released on the App Store and built the entire FastAPI, PostgreSQL, and Supabase backend, including synchronized workouts, ELO matchmaking, and weekly competition logic. His firsthand account also describes marketing before the UI existed, campus distribution, and why local community density mattered to the product.
This brief summarizes Karthik's public portfolio and firsthand accounts. Employment contributions and internal performance metrics are self-reported; linked code and artifacts provide additional evidence where available.