| name | resume-tailor |
| description | Tailors Harsha Vardhan Yellela's HTML resumes to specific job descriptions. Use when the user provides a job description and asks to create a tailored resume, make a resume for a specific role, or asks to "do the same" for a new JD. Outputs a tailored HTML resume to resumes/html/temp.html (or named file). Also handles updating existing resumes (sde, ml, devops, ai_automation, java_fullstack, resume.html).
|
Resume Tailor (FAANG 1-Page Standard)
CRITICAL: Resume Rules (Non-Negotiable)
- 1 page strictly — NO page breaks, NO 2-page resumes
- 2 experience entries ONLY — GRA (Jan 2025-Dec 2025) + Infor (Apr 2022-Dec 2023)
- NO CDF, NO Build Fellowship, NO EPAM — these are dropped from all resumes
- XYZ bullet format — "Accomplished [X] as measured by [Y] by doing [Z]"
- NO bold/strong inside bullet text — only in skill category labels
- 2-3 projects max with GitHub links, 2 bullets each
- Summary: 2 sentences, ~30 words — mirror JD language, no bold
- Skills: 4-5 categories, ~20 items total — only interviewable skills
- Education: degree + GPA only — NO coursework bullets
- NO Achievements section
Required Reading Before Tailoring
Always read these files first:
/home/har5ha/Desktop/resume/references/personal_details.md - education, experience, skills/home/har5ha/Desktop/resume/references/projects.md - full project list with descriptions/home/har5ha/Desktop/resume/CLAUDE.md - role-to-project mapping tables
Workflow
Step 1: Analyze the JD
Extract:
- Role type: ML/AI, SDE/Backend, DevOps, AI Automation, Full Stack, Java, etc.
- Must-have skills: technologies, tools, frameworks explicitly required
- Key phrases: exact language from the JD to mirror in the summary
- Nice-to-haves: bonus skills to include if Harsha has them
Step 2: Select Projects (2-3 max)
| Variant | Project 1 | Project 2 | Project 3 (if space) |
|---|
| SDE | Lambda Microservices (94 functions) | FieldFuze Backend (Go, RBAC) | Job Portal REST API |
| ML | Resume Optimizer (QLoRA, 9.5/10) | ML Sentiment Loop (MLOps) | LuffyGPT (from scratch) |
| DevOps | ML Sentiment Loop (Terraform, ECS) | Online Learning Portal (Jenkins, K8s) | Lambda Microservices |
| AI Automation | AgenticAI (CrewAI, LangChain, n8n) | CRE Research Agent (RAG) | Lambda Microservices |
| Java Full Stack | Inventra (Java 21, Spring Boot 3.5) | Job Portal REST API | FieldFuze Backend |
Step 3: Write the Summary
2 sentences, ~30 words, NO bold tags. Mirror JD verbs and nouns.
- BAD: "Experienced ML Engineer with Python skills"
- GOOD: "ML Engineer with experience fine-tuning LLMs (QLoRA/PEFT), building 8-microservice MLOps platforms, and deploying models on AWS SageMaker."
Step 4: Write Experience (2 entries only)
GRA Title by Role Type
| Role Type | GRA Title |
|---|
| SDE/Backend | Software Engineer Research Assistant – Agentic AI |
| ML/AI | ML Engineer Research Assistant – Agentic AI |
| DevOps | Platform Engineer Research Assistant – Agentic AI |
| AI Automation | AI Automation Research Assistant – Agentic AI |
| Java Full Stack | Software Engineer Research Assistant – Agentic AI |
Infor Title: Always SDE-1 (LN Technical Consultant)
Bullet Rules (XYZ Format)
- One bullet = one achievement, one metric
- Lead with impact verb (Reduced, Delivered, Deployed, Automated, Cut, Architected)
- Every bullet has at least one number (%, count, time, scale)
- NO semicolons — if it has one, split into separate bullets
- NO bold/strong tags inside bullet text
- GRA: 4 bullets, Infor: 3 bullets
GRA Base Bullets (reframe per role):
- Reduced manual research workflow time by 70% by designing multi-agent orchestration pipelines (CrewAI + LangChain) that automated literature review, data extraction, and report generation.
- Deployed 3 persistent agent services on AWS EKS/Fargate with OpenSearch Serverless for semantic search across 10K+ documents with sub-second retrieval.
- Cut infrastructure provisioning from 2 hours to 15 minutes by implementing Terraform IaC with GitHub Actions CI/CD for automated container deployments.
- [Role-specific bullet — ML: fine-tuning SLMs, DevOps: monitoring/Prometheus, AI: n8n benchmarking, SDE: FastAPI/GraphQL APIs]
Infor Base Bullets (same across all):
- Delivered REST API integrations for 3 enterprise clients (Ferrari, Boeing, Triumph) processing 500+ daily transactions through AWS Lambda with S3 event triggers and API Gateway.
- Reduced batch processing failures from weekly to monthly by resolving 15+ pipeline defects across MySQL databases and distributed ERP systems through systematic root-cause analysis.
- Cut deployment turnaround by 25% by containerizing business logic services with Docker and standardizing CI workflows across client environments.
Step 5: Write Skills (4-5 categories)
~20 items total. Only list what Harsha can discuss for 5 minutes in an interview.
DO NOT include: Jira, Confluence, IntelliJ IDEA, VS Code, GitBash, "Agile/Scrum", Claude Code, GitHub Copilot, OpenAI Codex
Step 6: Write Projects (2-3 entries, 2 bullets each)
- Include GitHub links in header
- NO bold in bullet text
- Each bullet should have at least one metric
Step 7: Output
- Copy CSS verbatim from
/home/har5ha/Desktop/resume/resumes/html/sde_engineer.html
- Write to
resumes/html/temp.html
- Tell user the suggested PDF name
CSS Reference (use exactly this)
body { margin: 0.3in auto 0.3in auto; line-height: 1.08; max-width: 7.5in; }
h1 { font-size: 22pt; margin: 6px 0 2px 0; }
h2 { font-size: 12pt; margin: 8px 0 2px; }
.contact-info { margin: 0 0 4px 0; }
p { margin: 2px 0 4px 0; }
.entry { margin-bottom: 5px; }
.entry .header { font-size: 10.5pt; }
.entry .subheader { font-size: 10.5pt; margin-bottom: 1px; }
ul { padding-left: 18px; font-size: 10pt; }
ul li { margin-bottom: 1px; }
ul li:before { left: -13px; }
Existing Resume Files
| File | Role |
|---|
resumes/html/sde_engineer.html | SDE/Backend/Full Stack (flagship) |
resumes/html/ml_engineer.html | ML/AI/Data Science |
resumes/html/devops_engineer.html | DevOps/Platform/SRE |
resumes/html/ai_automation_engineer.html | AI Automation/Agentic/Integration |
resumes/html/java_fullstack.html | Java Full Stack |
resumes/html/test_automation_enginner.html | QA/Test Automation |
resumes/html/temp.html | Scratch pad for one-off tailored resumes |
References
references/experience-angles.md - Per-role bullet patterns for each experience entry
