Resulo AI

I treat Resulo as a learning project. The goal was to work through how an agentic product decides which work is worth doing, how that decision is evaluated, and where a person has to stay in the loop.

Candidates spend significant time searching, evaluating fit, researching companies, tailoring resumes, writing application material, completing applications, tracking opportunities, and preparing for interviews.

More applications do not necessarily mean better applications.

Several products attempt to solve this by automatically applying to large numbers of jobs. I deliberately chose a different model. Application volume is not the same as application quality. Bot-based application systems also face increasing resistance from recruitment platforms attempting to detect and restrict automated activity.

Most importantly, there is one layer I do not believe AI should remove: human judgment.

The Resulo workflow is: source, then analyze, then qualify, then prepare, then review, then apply, then track. That is a deliberately different model from find, then auto-apply, then repeat.

Qualification happens before significant application effort is invested. The goal is better applications to better-matched opportunities, not maximum application volume.

Source: bring relevant opportunities into the job intelligence workflow.

Analyze: understand the role, company, requirements, responsibilities, seniority, skills, and relevant signals.

Qualify: compare the opportunity against the candidate's actual experience, skills, goals, and profile.

Prepare: generate a tailored application package using candidate and opportunity context.

Review: allow the candidate to review and edit consequential application material.

Apply: proceed after human review.

Track: maintain visibility into applications and progression.

I designed Resulo's job validation, research, and analysis methodology around strict qualification. The engine evaluates opportunities against the candidate rather than simply searching for keyword overlap. The system considers factors across candidate experience, skills, career direction, role requirements, company, seniority, opportunity quality, candidate preferences, and fit thresholds.

Users can further personalize elements of the qualification system. However, the product maintains minimum quality thresholds. Certain safeguards cannot simply be reduced until every available job appears to be a good match. This is intentional.

Job intelligence, qualification and fit scoring, tailored application preparation, candidate context, application tracking, interview intelligence, and human review.

Once an opportunity qualifies, Resulo can prepare the candidate's application package using their actual professional context, including experience, skills, CV or resume, portfolio, previous work, career objectives, role requirements, and company context. The candidate receives the prepared package and makes the final decision before submission.

When a candidate progresses to interview, Resulo changes modes. The system can prepare company intelligence, business model analysis, market context, role analysis, likely interview themes, potential questions, candidate-specific preparation, experience positioning, and negotiation strategy.

The product therefore supports the journey from opportunity discovery through interview preparation rather than ending at job sourcing.

During testing, the evaluation workflow analyzed 1,347+ jobs and produced 189 application packages. These are test counts from the learning project. They are not user numbers.

Primary AI: GPT-5.6 and Gemini. The product uses AI for different intelligence workloads including opportunity analysis, research, matching, qualification, synthesis, application preparation, company intelligence, and interview preparation.

The value is in the structured qualification methodology, workflow architecture, candidate context, thresholds, and orchestration around the models, not a generic LLM wrapper.

Automate the repetitive work. Preserve the candidate's judgment.

Human in the loop is intentional architecture. Resulo does not attempt to remove the candidate from their own job search. AI performs the expensive repetitive work. Human judgment remains responsible for the final representation of the candidate.

Resulo shows my approach to agentic product design: automation should not simply increase throughput. The product must understand when work is worth doing, prepare the work intelligently, and preserve human control at consequential decision points.