File size: 8,945 Bytes
9ae77d7
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
[
  {
    "id": "eval_001",
    "question": "Who is Arun Yadav and what does he do?",
    "category": "identity",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["Freelance AI Systems Engineer", "RAG Pipelines", "Autonomous Agents", "Data Scrapers"]
  },
  {
    "id": "eval_002",
    "question": "What is Arun's primary tech stack?",
    "category": "identity",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["Python", "FastAPI", "LangChain", "ChromaDB", "Playwright", "Asyncio"]
  },
  {
    "id": "eval_003",
    "question": "What are Arun's immediate career goals?",
    "category": "identity_goals",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["Freelance Consulting", "Solving High-ROI bottlenecks", "Building scalable AI systems"]
  },
  {
    "id": "eval_004",
    "question": "Did Arun go to college for Computer Science?",
    "category": "personal_background",
    "expected_source": "raw/personal_background.md",
    "expected_topics": ["Non-linear path", "NEET/JEE background", "B.Sc Degree", "Self-taught"]
  },
  {
    "id": "eval_005",
    "question": "How did Arun learn to code initially?",
    "category": "personal_background",
    "expected_source": "raw/personal_background.md",
    "expected_topics": ["PyDroid 3", "Smartphone", "Active building vs passive tutorials"]
  },
  {
    "id": "eval_006",
    "question": "How does Arun feel about theoretical, bookish learning?",
    "category": "personal_philosophy",
    "expected_source": "raw/personal_background.md",
    "expected_topics": ["Creates friction", "Accelerated independent learning", "Build-first execution"]
  },
  {
    "id": "eval_007",
    "question": "What is the Legal RAG System?",
    "category": "project_overview_legal",
    "expected_source": "github/legal_RAG_system/readme.md",
    "expected_topics": ["Indian legal documents", "IPC", "Constitution", "Exact-reference routing"]
  },
  {
    "id": "eval_008",
    "question": "How does the Legal RAG System handle document chunking?",
    "category": "project_tech_legal",
    "expected_source": "github/legal_RAG_system/architecture.md",
    "expected_topics": ["Document-aware structural chunking", "Statutes by section", "Constitution by article", "Judgments by paragraph"]
  },
  {
    "id": "eval_009",
    "question": "Why didn't Arun use generic text splitting for the Legal RAG project?",
    "category": "project_decisions_legal",
    "expected_source": "github/legal_RAG_system/decisions.md",
    "expected_topics": ["Destroys structural boundary", "Context loss", "Section integrity"]
  },
  {
    "id": "eval_010",
    "question": "Why does the Legal RAG System use a local ChromaDB instead of Pinecone?",
    "category": "project_decisions_legal",
    "expected_source": "github/legal_RAG_system/decisions.md",
    "expected_topics": ["Data privacy", "No cloud exposure for legal docs", "Local vector DB"]
  },
  {
    "id": "eval_011",
    "question": "What is the 99acres Real Estate Scraper Suite?",
    "category": "project_overview_scraper",
    "expected_source": "github/real_state_listing_scraper/readme.md",
    "expected_topics": ["3-track scraping suite", "Cloudflare bypass", "Structured CSV export"]
  },
  {
    "id": "eval_012",
    "question": "Why does Arun use ScraperAPI instead of building a custom proxy rotation?",
    "category": "project_decisions_scraper",
    "expected_source": "github/real_state_listing_scraper/decisions.md",
    "expected_topics": ["Cloudflare 403 Forbidden", "Built-in CAPTCHA solving", "Unreliable free proxies"]
  },
  {
    "id": "eval_013",
    "question": "In scraping, what is semantic anchoring and why does Arun use it?",
    "category": "project_decisions_scraper",
    "expected_source": "github/real_state_listing_scraper/decisions.md",
    "expected_topics": ["Anchoring on ₹ symbol", "Obfuscated CSS classes", "Robustness against UI changes"]
  },
  {
    "id": "eval_014",
    "question": "How does Arun prevent his scraper from being blocked when sending hundreds of requests?",
    "category": "project_tech_scraper",
    "expected_source": "github/real_state_listing_scraper/decisions.md",
    "expected_topics": ["asyncio.Semaphore", "Throttling", "Concurrency control"]
  },
  {
    "id": "eval_015",
    "question": "What is the 'personal_ai_agent' project?",
    "category": "project_overview_agent",
    "expected_source": "github/personal_ai_agent/readme.md",
    "expected_topics": ["Digital twin", "Tool-calling loop", "Telegram lead capture"]
  },
  {
    "id": "eval_016",
    "question": "How does Arun's personal AI agent handle rate limit failures?",
    "category": "project_tech_agent",
    "expected_source": "github/personal_ai_agent/decisions.md",
    "expected_topics": ["Multi-model fallback chain", "Groq models", "Graceful degradation"]
  },
  {
    "id": "eval_017",
    "question": "Why is the context for the personal AI agent injected from a plain text file?",
    "category": "project_decisions_agent",
    "expected_source": "github/personal_ai_agent/decisions.md",
    "expected_topics": ["Flexibility", "No code redeployment needed", "Separation of identity data"]
  },
  {
    "id": "eval_018",
    "question": "What happens if Arun's personal CLI agent cannot find the answer in the retrieved legal text?",
    "category": "project_tech_legal",
    "expected_source": "github/legal_RAG_system/code_summaries.json",
    "expected_topics": ["Graceful fallback", "Returns 'I do not have information regarding this'", "Zero hallucination"]
  },
  {
    "id": "eval_019",
    "question": "What are some of Arun's long-term visions?",
    "category": "identity_goals",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["Healthcare", "Education", "Real Estate", "Democratizing value"]
  },
  {
    "id": "eval_020",
    "question": "What does Arun's 'File Organiser' script do?",
    "category": "project_tier2",
    "expected_source": "github/neural_arun_labs/readme.md",
    "expected_topics": ["Auto-sorts into PDFs, Videos, Images, Others", "Daily-use utility"]
  },
  {
    "id": "eval_021",
    "question": "What did Arun learn from building the AI Snake game?",
    "category": "linkedin_posts",
    "expected_source": "linkedin/posts.md",
    "expected_topics": ["State management", "Game loop", "Human-in-the-loop UX"]
  },
  {
    "id": "eval_022",
    "question": "What is the 'Web Wizard' repository?",
    "category": "project_tier2",
    "expected_source": "github/web_wizard/readme.md",
    "expected_topics": ["Playwright curriculum", "Advanced web automation", "Distributed crawler systems"]
  },
  {
    "id": "eval_023",
    "question": "What is Arun's opinion on building 'AI Wrappers'?",
    "category": "personal_philosophy",
    "expected_source": "linkedin/profile_summary.md",
    "expected_topics": ["Avoids brittle wrappers", "Engineers robust solutions", "Systems engineering"]
  },
  {
    "id": "eval_024",
    "question": "How did Arun extract anomalies from the UPPCS examination results?",
    "category": "project_overview_anomaly",
    "expected_source": "github/result_anomaly/readme.md",
    "expected_topics": ["Regex extraction", "Roll numbers", "Series-prefix grouping"]
  },
  {
    "id": "eval_025",
    "question": "Why didn't Arun use OCR for the UPPCS result anomaly project?",
    "category": "project_decisions_anomaly",
    "expected_source": "github/result_anomaly/decisions.md",
    "expected_topics": ["pdfplumber", "Native text extraction", "High accuracy over OCR"]
  },
  {
    "id": "eval_026",
    "question": "What framework is Arun currently learning for multi-agent workflows?",
    "category": "identity_learning",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["LangGraph", "CrewAI", "AutoGen", "MCP"]
  },
  {
    "id": "eval_027",
    "question": "What is Arun's work environment like when he's deeply focused?",
    "category": "personal_background",
    "expected_source": "raw/personal_background.md",
    "expected_topics": ["Strictly minimal", "Laptop, keyboard, notebook"]
  },
  {
    "id": "eval_028",
    "question": "How much money does Arun make?",
    "category": "negative_test",
    "expected_source": "static/rules_of_engagement.md",
    "expected_topics": ["Polite refusal", "Out of bounds", "No hallucination"]
  },
  {
    "id": "eval_029",
    "question": "Where does Arun live exactly?",
    "category": "negative_test",
    "expected_source": "static/rules_of_engagement.md",
    "expected_topics": ["Prayagraj, UP, India", "No exact address provided"]
  },
  {
    "id": "eval_030",
    "question": "Can Arun set up a full-stack Next.js app for me?",
    "category": "negative_test_tech",
    "expected_source": "static/public_profile.md",
    "expected_topics": ["Backend focused", "Python, FastAPI", "AI integrations", "Does not do frontend/Next.js currently"]
  }
]