The Sahil Bachu
Vol. I · No. 579Rochester, New York, Thursday, September 24, 2026Agentic AI · Page B23
All workSection B · Page B23

Agentic AIDay-by-Day

A Life Coach Agent That Remembers Your Excuses

A tough-love AI coach that breaks goals into daily tasks, remembers past conversations and pushes back when you keep rescheduling.

By Sahil BachuAgentsPrototype

Graphic5 steps
  1. PlanLLM drafts milestones and daily tasks as JSON
  2. ReviewYou approve or ask for changes
  3. TrackTasks completed or failed each day
  4. RememberChromaDB memory feeds every prompt
  5. RescheduleCascade the rest of the plan, with warnings
What happens inside Day-by-Day, in order.Graphic: The Sahil Bachu

Rochester, N.Y. A to-do app will let you push a task back forever without comment. Day-by-Day keeps count. Give it a goal and a deadline and it drafts weekly milestones and daily tasks for you to review. Once you confirm, it tracks what you finish and what you skip, and its strict-but-fair persona starts warning you on your second and third reschedule in a month.

The backend is FastAPI with a LangChain ReAct agent on Groq through LiteLLM. It has about 19 tools: create and break down goals, complete or fail tasks, search past conversations, analyze failures and cascade-reschedule the rest of a plan. Memory lives in ChromaDB, with conversations, goals and tasks embedded by MiniLM, so every prompt gets the relevant history. When Groq rate-limits the 8B model, it falls back to 70B, then to a plain memory-aware reply.

The Next.js front end has chat, today's tasks, a goals dashboard, a plan review flow and @goal mentions. Behind the chat, the bulk of the code is plumbing: the model fallback chain, the memory layer and the scheduler that spreads tasks up to the deadline.

More from The Work

See all 25

Multi-agent

Hire AI Employees, Then Walk Over to Their Desks

A pixel-art office where every AI agent is an employee with its own desk, skills, scheduled jobs and Docker container.

PixelHQ turns the open-source SkyOffice into a single-player office for AI agents. Walk up to an employee, chat with it, give it skills and scheduled jobs, and feed a shared knowledge base. Each employee runs in its own container.

Archived · 2026

Continued on Page B17

Agentic AI

Your Coding Agent Fills the Gaps in Your Resume

A local-first orchestrator that reads a student's resume, proposes projects to fill the gaps and has their own Claude Code or Codex build and deploy them.

ResumeGod reads your resume, suggests projects that would fill its gaps, then drives your own Claude Code or Codex to build and deploy them. It finishes with a personal site and a fresh LaTeX resume. Nothing is hosted centrally.

Prototype · 2026

Continued on Page B15

Generative UI

A Canvas of Widgets You Order in Plain English

An infinite canvas where an AI agent turns a plain-English request into a live widget wired to your own data.

Ask for something like a card that summarizes your unread emails, and an agent researches, designs and places a live widget on an infinite, zoomable grid. Connectors for Gmail, Calendar, Slack, GitHub, Linear and Notion feed it real data.

In progress · 2026

Continued on Page B11