Currently: building Unni Open to: founder's office roles
byananya.

Ananya,
AI automation builder.operations generalist.web designer.runner.

Worked with teams at

Mid Touch Media, FranceAssist My Day, CanadaRio ImmigrationConsortium StaffingsMid Touch Media, FranceAssist My Day, CanadaRio ImmigrationConsortium Staffings

Two years running operations for small remote teams taught me one rule: if you do it twice, automate it. So I started building the systems. Then the AI.

Work.

Things I've built, what they do, and what I learned building them.

Showing 5 of 5 projects

Clause. Teaching AI which number on a datasheet is the right one.

  • #AI
  • #Evaluation
  • #API
Live API
99.4%field accuracy
Type
AI extraction service
Year
2026
Role
Solo builder
Outcome
161 of 162 fields right, score predicted before the run, $0 spent.
  • Domain rules, not parsing.

    Datasheets print two plausible numbers for the same question. Knowing which one is right is the real work.

  • Measured, not guessed.

    Hand-written ground truth, measured run-to-run variance, and a prediction committed before the baseline.

  • Built, tested, removed.

    Citations, retrieval and a verification pass were each measured, then cut when the numbers didn't justify them.

  • Production basics.

    Auth, rate limits, request IDs and a 15-check API test suite, all on a free tier.

agent-contract. Catching AI agents that fail politely.

  • #AI
  • #n8n
  • #OpenSource
v0.3, open source
Type
n8n Code node, open source
Year
2026
Role
Solo builder
Outcome
Nine checks, 14 test fixtures, verified in a live n8n workflow.
  • Green isn't working.

    n8n marks an AI node successful if it doesn't crash. A model always returns text, so failures look like success.

  • Phantom tool calls.

    Flags agents that answer without calling the tool they were told to use.

  • No AI checking AI.

    Plain string and JSON checks: offline, instant, free, and the same verdict every time.

  • Shaped by the community.

    Two features came straight from feedback in the n8n community.

Founder outreach. Eight founders researched before breakfast.

  • #Automation
  • #AI
Running daily
Type
n8n automation
Year
2026
Role
Solo builder
Outcome
Runs every day on free-tier tools: sourcing, research, briefs and reminders.
  • Weekly sourcing.

    Pulls India-based YC companies and Series A deals from startup news.

  • Morning research.

    Three web searches per company, then AI writes a short brief on each founder.

  • Local first.

    Kerala-based companies rise to the top, then the newest Series A rounds.

  • Nudges on my phone.

    A morning list, an evening check-in, and follow-up reminders when someone goes quiet.

Unni. Skyscanner, but for clothes.

  • #AI
  • #Product
In development
Type
AI shopping assistant
Year
2026
Role
Founder and builder
Status
Building now, testing with friends next.
  • Photo in.

    Snap a dress, shoe or bag and Unni finds where it's from.

  • Price comparison.

    The same item, compared across sites.

  • US availability.

    Whether it's sold in the US, and how to get it here.

  • Lookalikes.

    Three to five cheaper alternatives, with images and links.

Crypto data pipeline. Built to survive the bad days.

  • #Data
  • #Pipelines
Complete
Type
Data pipeline
Role
Solo builder
Outcome
A production-style setup built for reliability and data consistency.
  • Orchestrated.

    Airflow schedules each step and handles retries.

  • Consistent.

    Docker and PostgreSQL keep it reproducible, even when a run fails halfway.

About.

I also run, play tennis, badminton and padel, and write about all of it.

I'm Ananya. I studied chemical engineering, then spent two years in operations and web design for remote teams in France and Canada. Somewhere along the way I fell for automation, and then for AI.

Away from the screen, you'll find me on a run or on a court.

  • Operations SpecialistMid Touch Media, France (remote)
  • Web Developer and Operations AssociateAssist My Day, Canada (remote)
  • Web Developer InternConsortium Staffings and Solutions, Canada (remote)
  • BE, Chemical EngineeringSiddaganga Institute of Technology

Stack.

What I reach for, day to day.

Blog.

A post every week on AI, automation and software, and a monthly running log.

Let's build something that actually works.

Founder's office, operations, or an automation you're tired of doing by hand. Tell me about it.