SShortSingh.
Back to feed

init.dev: Offline AI Career Co-Pilot for Roadmaps, Resume Scoring and Mock Interviews

0
·2 views

init.dev is an open-source, offline-first career assistant built to run locally on low-resource machines without a GPU. It parses a user's career goals and current skills to generate a structured learning roadmap. The resume scoring feature estimates ATS pass probability, using TabPFN-3.5 as the primary model with a local Gemma fallback. The tool also includes a persistent study chat and realistic role-specific mock technical interviews run in the terminal. Its developer says local processing avoids API costs and keeps resumes and learning histories off closed cloud services.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

Log in to join the discussion and vote.

Log in

Related stories

0
ProgrammingDEV Community ·

GST Helper: Local-First Tool for Preparing Selected GSTR-1 Values from Invoices

GST Helper is an open-source, local-first app that reads outward-sales invoice PDFs to help small businesses prepare selected GSTR-1 values. It covers regular B2B invoices in Table 4A, the B2B HSN/SAC summary in Table 12, and documents issued in Table 13, with page evidence shown for each extracted value. The tool produces a reference PDF tied to the saved review, so users can check figures before entering them on the GST Portal. It does not file returns, connect to the portal, or prepare the full GSTR-1, and other sections must be handled separately. The project was built as a submission to the Hacktoberfest Open-Source AI Challenge and is available on GitHub.

0
ProgrammingDEV Community ·

Designing a Control Plane for an AI-Run Firm with Workflow, Agent and Human Lanes

A DEV Community post argues that an 'autonomous company' is really a control-plane design rather than a matter of model choice. It illustrates the idea with a hypothetical small service firm handling lead intake, with no real results reported. Tasks are divided into three lanes: fixed workflows, agent-driven steps within limits, and human decisions, with the lane enforced in code. The author notes that tool access, approval requirements and logging are set in code rather than in prompts, and that a named owner must be able to pause any task. Money, contracts and customer promises are always routed to a person, and the post concludes that such controls are what make human oversight of AI operations real.

0
ProgrammingDEV Community ·

Bigger LLM Coordinator Fails to Fix Agent Loop; Deterministic Script Does

A series installment reports that upgrading to a larger coordinator model did not resolve orchestration failures such as dropped steps, delegations emitted as text, and empty turns after tool calls. The author argues the root cause is the agentic loop itself, since the model must simultaneously plan, dispatch, track outstanding work, and evaluate results. Handing the control loop to a small deterministic script, with the model used only for planning and evaluation, produced immediate gains. In the reported test, a plan that had stalled after three of five steps completed fully once executed by the script.

0
ProgrammingDEV Community ·

One-Time Payment or Subscription? How Vibe-Coded App Makers Should Decide

A solo developer in Hyderabad explains how creators of apps built with AI coding tools like Claude Code, Cursor, and Codex can choose between charging once and charging recurring fees. The core test is whether each user costs the developer money to serve, since usage-driven costs such as AI model calls or hosting favor subscriptions or credits. One-time pricing suits low-cost tools, templates, and downloadable products, though it requires a steady flow of new buyers to keep earning. Subscriptions offer predictable monthly income but add complexity around cancellations, failed payments, and plan changes. The guide also outlines the basic setup for each model, including checkout links, payment notifications, and access controls.