Siri Rama
An economist by training and a storyteller by nature, I'm building at the intersection of AI, GTM & technology.
Toronto-based, working in business development across AI agents, go-to-market strategy, and enterprise technology — informed by a background in economics and politics. The projects below are the build side of that: agents, automation, and open-data tools, most of which started as something I needed and kept going.
The case files.
Tools I built to get from raw data to a story worth telling — agents, automation, and analysis. Each one shipped; each one taught me something the docs didn't. Live demos hang off subdomains of this site as they go up.
Talk to the file.
A live voice agent, grounded in my actual resume and project files. Ten minutes, no fabrication.
On-Demand Voicemail Generator
A bilingual (EN/FR) tool that turns a prospect's name, title, and company into a playable voicemail — in a cloned voice — in about ten seconds, drafted by Claude and synthesized by ElevenLabs. The same three-stage pipeline (draft → synthesize → deliver) runs from a live form mid-call or in batches over a CSV, and it forks at the one point that matters: playback is instant and gate-free, but sending it to someone else always stops for an explicit confirmation showing exactly who, what, and how large first. Built as a tested Python CLI, then ported to Next.js API routes so it could run live on the site instead of needing its own server.
Read the case file →Account Dossier Agent
A multi-agent pipeline that turns a company's site, blog, job board, SEC filings, and news coverage into a cited research dossier — then drafts outreach that's only allowed to reference what the dossier actually verified. Every claim carries a verbatim quote checked programmatically against its source; if the model can't quote it exactly, the claim gets rejected, not silently invented. Corroboration is tiered — site content and ATS job data trust first, SEC filings second, GDELT news third — so a single stray headline never gets treated as verified fact.
Read the case file → · View source →Lead Qualification Agent
A lead-qualification pipeline built to solve the problem I run into at Tipalti every week: inbound funnels drown reps in volume, and the real buyers go stale while triage takes days. It resolves messy company names against real filings data, scores leads against an auditable SQL model a RevOps lead could read and retune without touching code, and reserves the LLM for the one job only it's suited for — reading intent out of an ambiguous message. On synthetic test data: 99.7% company-resolution accuracy, 99.1% precision and 100% recall on high-priority leads, and a clean 10/10 against adversarial prompt-injection attempts.
Read the case file → · View source →Executive Agent
An agentic assistant that reads across the messy inputs of a working day — email, calendar, notes, documents — and produces a single prioritized brief: what needs a decision, what can wait, and what to say. Built around a plan-and-review loop so the agent proposes, and a human approves, before anything leaves the room.
View source →Downloads Butler
A five-stage pipeline — scan, analyze, report, approve, execute — that inventories files with SHA-256 hashing, flags duplicates and stale installers with a deterministic rules engine plus an optional local-LLM pass, then presents an HTML approval page. Approved actions run through a quarantine system with a full restore path, so no file is ever destructively deleted. Built and adversarially tested against path-traversal and protected-file edge cases.
View source →ATIP Automation In progress
Tooling that takes the grind out of Access to Information requests — drafting, filing, and tracking.
This one's still under construction. Contact me for more info.
More experiments live on GitHub, including the ones that didn't make the cut.
The numbers make the case. The story wins the room.
I'm an economist by training and a storyteller by nature. Those usually pull in opposite directions — rigor versus persuasion — and most of my work is spent proving they don't have to.
I've used that combination on two very different audiences: voters, where data becomes canvass strategy and a message that moves turnout; and buying committees, where the same discipline becomes a business case a room of skeptics will sign off on. Different stakes, identical method — find what the data actually says, then make people feel why it matters.
The projects on this site are the engine room behind that. When the analysis got repetitive, I automated it; when the data was locked in a portal, I wrote something to free it. Everything on the technical side is self-taught — mostly by breaking things and reading the error messages carefully.
The storyteller
- Data storytelling
- GTM Strategy
- B2B sales narratives
- Executive Briefs
- Stakeholder communication
The economist
- Economic & statistical analysis
- Python
- SQL and business intelligence analysis
- LLM & agentic workflows (Ollama)
- Open-data pipelines
Off the clock: meditation, tea, politics, history, strategy games, and the occasional side project that starts as a one-off script and refuses to stay small.