Clean Repo Initialiazed with generator

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+app/static/fonts/Roboto-Regular.ttf filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,14 @@
+FROM python:3.11-slim
+
+WORKDIR /app
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+COPY . .
+
+# Create cache folder and fix permissions
+RUN mkdir /.cache
+RUN chmod -R 777 /.cache
+
+RUN mkdir -p /code/backend/transformers_cache && chmod -R 777 /code/backend/transformers_cache
+
+CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -0,0 +1,10 @@
+---
+title: Resume Parser
+emoji: 😻
+colorFrom: yellow
+colorTo: yellow
+sdk: docker
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app/api/routes.py

@@ -0,0 +1,24 @@
+from fastapi import APIRouter, HTTPException
+from app.models.schema import GenerateRequest, GenerateResponse
+from app.services.generator import generate_cover_letter
+from app.services.pdf_creator import save_pdf
+from app.services.resume_parser import extract_text_from_resume
+from app.utils.file_utils import generate_unique_filename
+
+router = APIRouter()
+
+@router.post("/generate", response_model=GenerateResponse)
+async def generate_cover_letter_api(data: GenerateRequest):
+    try:
+        resume_text = extract_text_from_resume(data.resume_path)
+        letter_text = generate_cover_letter(data, resume_text)
+
+        filename = generate_unique_filename()
+        pdf_path = save_pdf(letter_text, filename)
+
+        return GenerateResponse(
+            letter=letter_text,
+            pdf_url=f"/{filename}"
+        )
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

app/main.py

@@ -0,0 +1,45 @@
+from fastapi import FastAPI, Request
+from fastapi.staticfiles import StaticFiles
+from app.api.routes import router
+import os
+import logging
+
+logging.basicConfig(level=logging.INFO)
+app = FastAPI(title="Cover Letter Generator")
+
+BASE_DIR = os.path.dirname(os.path.abspath(__file__))
+
+
+
+# Serve resumes (these are static, bundled in repo, read-only is fine)
+app.mount(
+    "/resumes",
+    StaticFiles(directory=os.path.join(BASE_DIR, "static/resumes")),
+    name="resumes"
+)
+
+app.mount("/fonts", StaticFiles(directory=os.path.join(BASE_DIR, "static/fonts")), name="fonts")
+
+# Writable directory for PDFs
+PDF_DIR = "/tmp/pdfs"
+os.makedirs(PDF_DIR, exist_ok=True)
+
+# Serve PDFs (generated at runtime) under /static/pdfs
+app.mount("/static/pdfs", StaticFiles(directory=PDF_DIR), name="pdfs")
+
+
+@app.middleware("http")
+async def log_requests(request: Request, call_next):
+    logging.info(f"API hit: {request.method} {request.url}")
+    response = await call_next(request)
+    return response
+
+
+@app.get("/")
+def ping():
+    print()
+    return {"status": "ok"}
+
+
+# Register routes
+app.include_router(router)

app/models/schema.py

@@ -0,0 +1,9 @@
+from pydantic import BaseModel
+
+class GenerateRequest(BaseModel):
+    job_details: str
+    resume_path: str
+
+class GenerateResponse(BaseModel):
+    letter: str
+    pdf_url: str

app/services/generator.py

@@ -0,0 +1,130 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
+import torch
+from app.models.schema import GenerateRequest
+import os
+from fastapi import FastAPI, HTTPException
+
+
+os.environ["TRANSFORMERS_CACHE"] = "/code/backend/transformers_cache"
+os.makedirs(os.environ["TRANSFORMERS_CACHE"], exist_ok=True)
+
+
+# Initialization model immediately after server is Up!
+device = "cuda" if torch.cuda.is_available() else "cpu"
+dtype = torch.float16 if device == "cuda" else torch.float32
+
+# TinyLlama/TinyLlama-1.1B-Chat-v1.0
+model_name = "Qwen/Qwen2.5-0.5B-Instruct"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    torch_dtype="auto",
+    device_map="auto"
+)
+
+model = model.to(device)
+
+
+def generate_cover_letter(data: GenerateRequest, resume_text: str) -> str:
+    prompt = f"""
+AI Cover Letter Writer Bot Prompt
+
+You are an expert cover letter writer that creates personalized, human-written cover letters based on job descriptions and CV information. Your goal is to craft authentic letters that sound like they're written by a Pakistani professional with 6.5 IELTS band score - natural, clear, but not overly sophisticated or AI-generated.
+
+Input Requirements
+
+- **Job Details:** Complete job posting with requirements, responsibilities, and company information
+Job Details: {data.job_details}
+
+- **CV/Resume:** Tailored resume specifically prepared for the target role
+Resume Text: {resume_text}
+
+Output Format Required:
+Dear [Hiring Team/Company Name extracted from Job Details],
+
+[Paragraph 1: Position + why interested in company]
+[Paragraph 2: Relevant experience + skills matching job requirements]  
+[Paragraph 3: Value you bring + enthusiasm to join]
+
+Best regards,
+- Name of user extract from resume
+
+**Do not include the following data from resume:**
+
+- Header with personal contact information
+- Date
+- Subject line
+- Closing signature
+- Any formatting beyond the basic greeting and paragraphs
+
+Rules:
+- 3 paragraphs only (50-80 words each)
+- Total: 200-300 words
+- Sound natural, not AI-written
+- Match resume skills to job requirements
+- Mention specific technologies/companies from inputs
+- Use simple language, avoid: robust, leverage, utilize, optimal, innovative
+
+**Missing Skills? Say: "I am eager to learn [skill] and have quickly mastered [example] before"
+**Less Experience? Focus on quality: "My experience with [skill] includes [specific achievement]"
+
+Before finalizing, ensure the cover letter:
+
+- [ ] Addresses the specific role and company by name
+- [ ] Highlights 2-3 most relevant experiences from the CV
+- [ ] Addresses any obvious skill gaps with learning commitment
+- [ ] Uses natural, varied language (avoiding the banned word list)
+- [ ] Uses simple, clear language that sounds human-written (not AI-generated)
+- [ ] Avoids overly sophisticated vocabulary or complex sentence structures
+- [ ] Includes a clear call to action
+- [ ] Stays within 3-4 paragraph limit
+
+Remember: Your goal is to write a cover letter that sounds like it was written by a real person with good English skills, not by AI. Keep it natural, straightforward, and genuine.
+"""
+    try:
+        
+        messages = [
+    {"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful AI covert letter bot."},
+    {"role": "user", "content": prompt}
+]
+
+        # Apply chat template
+        text = tokenizer.apply_chat_template(
+            messages,
+            tokenize=False,
+            add_generation_prompt=True
+        )
+
+        # Tokenize input
+        model_inputs = tokenizer(
+            [text],
+            return_tensors="pt",
+            padding=True,
+            truncation=True,
+            max_length=2048
+        ).to(model.device)
+
+        # Generate with better parameters
+        generated_ids = model.generate(
+            **model_inputs,
+            max_new_tokens=512,
+            do_sample=True,           # sampling enabled for more natural output
+            temperature=0.7,          # control creativity
+            top_p=0.9,                # nucleus sampling
+            top_k=50,                 # top-k sampling
+            repetition_penalty=1.1,   # reduce repetition
+            eos_token_id=tokenizer.eos_token_id
+        )
+
+        # Decode output
+        response = tokenizer.decode(
+            generated_ids[0], 
+            skip_special_tokens=True
+        )
+
+        return response
+
+
+
+    except Exception as e:
+        return {"error": str(e)}

app/services/pdf_creator.py

@@ -0,0 +1,46 @@
+from fpdf import FPDF
+import os
+from datetime import datetime
+
+PDF_DIR = "/tmp/pdfs"
+os.makedirs(PDF_DIR, exist_ok=True)
+
+# fallback for local testing
+SPACE_URL = os.getenv("SPACE_URL", "http://localhost:8000")
+
+
+def normalize_text(text: str) -> str:
+    replacements = {
+        "’": "'", "‘": "'",
+        "“": '"', "”": '"',
+        "—": "-", "–": "-",
+        "…": "...",
+    }
+    for old, new in replacements.items():
+        text = text.replace(old, new)
+    return text
+
+
+def save_pdf(text: str, filename: str) -> str:
+    text = normalize_text(text)
+    path = os.path.join(PDF_DIR, filename)
+
+    pdf = FPDF()
+    pdf.add_page()
+    pdf.set_font("Arial", size=11)  # Built-in Latin-1 font
+    pdf.set_auto_page_break(auto=True, margin=15)
+
+    sections = text.split('\n\n')
+    for section in sections:
+        if section.strip():
+            if any(word in section.lower() for word in ["best regards", "yours sincerely", "sincerely"]):
+                pdf.ln(5)
+                lines = section.strip().split('\n')
+                for line in lines:
+                    pdf.cell(0, 6, line.strip(), ln=True, align='L')
+            else:
+                pdf.multi_cell(0, 6, section.strip(), align='L')
+                pdf.ln(8)
+
+    pdf.output(path)
+    return f"{SPACE_URL}/static/pdfs/{filename}"

app/services/resume_parser.py

@@ -0,0 +1,8 @@
+import pdfplumber
+
+def extract_text_from_resume(file_path: str) -> str:
+    text = ""
+    with pdfplumber.open(file_path) as pdf:
+        for page in pdf.pages:
+            text += page.extract_text() or ""
+    return text.strip()

app/utils/file_utils.py

@@ -0,0 +1,5 @@
+import uuid
+
+def generate_unique_filename(extension: str = "pdf") -> str:
+    """Generate a unique filename with given extension."""
+    return f"{uuid.uuid4().hex}.{extension}"

requirements.txt

@@ -0,0 +1,8 @@
+fastapi
+uvicorn
+transformers
+torch
+fpdf
+pdfplumber
+accelerate
+safetensors