diff --git a/group-solution.py b/group-solution.py
new file mode 100644
index 0000000..5fe16e5
--- /dev/null
+++ b/group-solution.py
@@ -0,0 +1,50 @@
+group = {
+    "Jill": {
+        "age": 26,
+        "job": "biologist",
+        "relations": {
+            "Zalika": "friend",
+            "John": "partner"
+        }
+    },
+    "Zalika": {
+        "age": 28,
+        "job": "artist",
+        "relations": {
+            "Jill": "friend"
+        }
+    },
+    "John": {
+        "age": 27,
+        "job": "writer",
+        "relations": {
+            "Jill": "partner"
+        }
+    },
+    "Nash": {
+        "age": 34,
+        "job": "chef",
+        "relations": {
+            "John": "cousin",
+            "Zalika": "landlord"
+        }
+    }
+}
+
+# Saving structure to JSON
+import json
+
+print(json.dumps(group))
+
+with open('group-data.json', 'w') as output:
+    output.write(json.dumps(group))
+
+# Reading json
+group_reload = None
+with open('group-data.json', 'r') as input:
+    group_reload = input.read()
+
+group_reload = json.loads(group_reload)
+print(group_reload)
+print(type(group_reload))
+assert group_reload == group
\ No newline at end of file
diff --git a/group.py b/group.py
index e2ec347..c6ca97c 100644
--- a/group.py
+++ b/group.py
@@ -1,5 +1,79 @@
 """An example of how to represent a group of acquaintances in Python."""
-
+# Testing push/pull
 # Your code to go here...
 
-my_group =
+my_group = {
+    # (name, age) as key
+    # ["job", list of relations as tuples (name, relation to name)] as value
+    ("Jill", 26):["biologist", ("John", "partner"), ("Zalika", "friend")], 
+    ("Zalika", 28):["artist", ("Jill", "friend")],
+    ("John", 27):["writer", ("Jill","partner")],
+    ("Nash", 34):["cook", ("John", "cousin"), ("Zalika","landlord")]
+}
+
+# Flexibly dataset to add people with various relations
+# Difficulty in searching 
+# It allows people with no connections
+# Doesn't assume reciprocal relations
+
+# Find maximum age
+max_age, max_age_person = None, None
+people = [x[0] for x in my_group.keys()]
+ages = [x[1] for x in my_group.keys()]
+max_age = max(ages)
+max_age_people = people[ages.index(max_age)]
+
+print('Find maximum age')
+print(max_age, max_age_people)
+
+# Find mean no. connections in the group
+mean_no_connections = None
+no_connections = [len(x[1:]) for x in my_group.values()]
+mean_no_connections = sum(no_connections)/len(my_group.keys())
+print('Find mean no. connections')
+print(mean_no_connections)
+
+# Find maximum age with at least one relation
+
+max_age2, max_age_person2 = None, None
+
+# Find people with at least one connection
+people_conditioned = [person for person in people if no_connections[people.index(person)]>0]
+# Filter ages according to one-connection conditions
+ages_conditioned = [age[1] for age in my_group.keys() if age[0] in people_conditioned ]
+#print(people_conditioned, ages_conditioned)
+max_age2 = max(ages_conditioned)
+max_age_person2 = people[ages_conditioned.index(max_age2)]
+print('Find people with at least one relation')
+print(max_age2, max_age_person2)
+
+# Find maximum age with at least one relation
+
+max_age3, max_age_person3 = None, None
+
+# Find relations for each person
+relations = [my_group[person][1:] for person in my_group.keys()]
+#print(relations)
+
+# Find no_friends
+no_friends = relations
+for index, relation in enumerate(relations):
+    no_friends[relations.index(relation)] = 0
+    for relation_sub in relation:
+        if relation_sub[1] == 'friend':
+            no_friends[index] += 1
+
+#print(no_friends)
+# Find people with at least one friendship
+
+people_conditioned2 = [person for person in people if no_friends[people.index(person)]>0]
+
+# Filter ages according to one-friendship conditions
+
+ages_conditioned2 = [age[1] for age in my_group.keys() if age[0] in people_conditioned2 ]
+
+max_age3 = max(ages_conditioned2)
+max_age_person3 = people[ages_conditioned2.index(max_age3)]
+print('Find people with at least one friend')
+print(max_age3, max_age_person3)
+