JobAgent

Upload resume view and we take care of the rest view
JobAgent User view
AGI agent applying to jobs and filling out form view
Job detail view with the connection messages.
Practice Interview with Telnyx View.
Resume Extraction View.

Inspiration

Searching and applying for jobs is repetitive, time-consuming, and often discouraging. Candidates re-enter the same information into countless forms, manually tweak similar cover letters, and struggle to figure out where and how to stand out. We wanted to build an AI assistant that feels like a dedicated job search co-pilot — one that can handle the busywork of applications, help identify the right roles, surface networking opportunities, and even prepare users for interviews.

What it does

JobAgent is an AI-powered job application assistant that automates the end-to-end application process using the AGI API for browser automation.

Here’s the user journey:

Resume Upload
Users start by uploading their resume into JobAgent.
Resume Analysis
The agent analyzes the resume to understand the user’s background, skills, and experience.
Job Search on NetworkIn
Using that profile, JobAgent searches for relevant jobs on NetworkIn, a LinkedIn-like platform.
Job Title Recommendations
It recommends job titles that align with the user’s background so the user can choose which role(s) to target.
Company Selection
After a job title is selected, JobAgent identifies suitable companies hiring for that role and waits for the user to pick where they want to apply.
Tailored Cover Letters
For each chosen company, the agent generates a tailored cover letter and returns it to the user for review and approval.
Automated Application Submission
Once a cover letter is approved, JobAgent uses browser automation to complete and submit the application form on the user’s behalf.
Networking & Referrals
The agent then searches for first- and second-degree connections at the selected company, suggesting potential contacts for coffee chats or referrals. The user chooses whom to reach out to, and JobAgent facilitates the process.
Mock Interview Preparation
After the application stage, users can schedule a mock interview. By entering their phone number and sending a request, a Telnyx AI agent conducts a practice interview, using the specific company and job title to ask realistic, contextual questions.

How we built it

JobAgent is composed of several coordinated components:

AGI API Browser Automation
We use the AGI API to navigate job listings and application forms on NetworkIn and company career pages. It handles actions like logging in, clicking through job postings, filling in fields, and submitting applications.
Resume Parsing & Profile Understanding
The resume is parsed into structured data (skills, roles, experience, education). This structure powers job search queries on NetworkIn and informs the job title recommendations.
Job Search & Recommendation Logic
Based on the parsed profile, JobAgent constructs targeted searches on NetworkIn, filters results for relevance, and recommends job titles aligned with the user’s background and seniority level.
Cover Letter Generation
For each company and role, the system combines the resume content with job description context to generate a tailored cover letter that highlights relevant experience and alignment with the role.
Connection Discovery
JobAgent queries NetworkIn’s graph-like data to find first- and second-degree connections at the selected companies, ranking them as potential coffee chat or referral targets.
Telnyx AI Integration
When the user requests a mock interview, we pass the company and role context, plus key resume highlights, to a Telnyx AI agent. It then conducts a phone-based practice interview using realistic questions tied to that specific role.

Challenges we ran into

Dynamic Job and Application Pages
Job listings and application forms often change layouts, add pop-ups, or introduce new fields. Making the browser automation robust and resilient to these changes was a core challenge.
High-Quality, Non-Generic Cover Letters
Generating tailored cover letters that feel specific and human, rather than generic AI text, required careful prompt and context design.
Connection Relevance
Not every contact at a target company is equally useful. Filtering and ranking connections so that suggestions feel genuinely helpful (e.g., shared background, relevant team) was more complex than simply listing everyone.
Context Synchronization for Mock Interviews
Ensuring that Telnyx AI always has up-to-date context (company, job title, and user profile) so mock interviews feel realistic and targeted required thoughtful data flow between components.

Accomplishments that we're proud of

True End-to-End Automation
JobAgent doesn’t just help with one step; it covers the entire pipeline from resume upload to final application submission, with user approvals at key checkpoints.
Integrated Networking Support
Going beyond traditional application helpers, JobAgent actively surfaces first- and second-degree connections and helps users reach out for coffee chats and referrals.
Contextual Mock Interviews
The Telnyx AI integration transforms interview prep from generic Q&A into realistic, job-specific practice sessions tied directly to applications the user has just submitted.
User-in-the-Loop Design
Users remain in control of crucial decisions — job titles, target companies, and final cover letters — while still benefiting from heavy automation on the repetitive tasks.

What we learned

Automation Needs Clear Approval Points
Users are much more comfortable with automation when they can review and confirm important steps, like which companies to apply to and what gets sent on their behalf.
Networking Multiplies the Value of Applications
Automatically pairing applications with suggested contacts for coffee chats and referrals makes the assistant feel more like a career coach than just a form filler.
Context Makes Practice Interviews More Valuable
Mock interviews that know the job title and company are significantly more useful than generic practice sessions, leading to better preparation and confidence.
Robust Web Automation Is All About Edge Cases
Small changes in web layouts and application flows can break naïve automation, so handling edge cases gracefully is essential for reliability.

What's next for JobAgent

Support for More Platforms
Expand beyond NetworkIn to additional job boards and applicant tracking systems so users can apply across the wider job market.
Application Tracking & Insights
Add a dashboard where users can track all their applications, see statuses, and gain insight into response rates and outcomes.
Deeper Personalization
Incorporate user preferences like location, company size, remote vs. onsite, and industry focus even more deeply into job search and recommendations.
Richer Interview Coaching
Build feedback summaries and personalized prep plans from Telnyx mock interviews, helping users focus on their weak spots.
Outreach Template Library
Offer a growing library of customizable outreach templates for coffee chats and referrals, tuned by user feedback over time.